+

WO2003085494A2 - Systeme de decodage video comprenant un decodeur a longueur variable programmable - Google Patents

Systeme de decodage video comprenant un decodeur a longueur variable programmable Download PDF

Info

Publication number
WO2003085494A2
WO2003085494A2 PCT/US2003/009896 US0309896W WO03085494A2 WO 2003085494 A2 WO2003085494 A2 WO 2003085494A2 US 0309896 W US0309896 W US 0309896W WO 03085494 A2 WO03085494 A2 WO 03085494A2
Authority
WO
WIPO (PCT)
Prior art keywords
variable
decoding
length
code
macroblock
Prior art date
Application number
PCT/US2003/009896
Other languages
English (en)
Other versions
WO2003085494A3 (fr
Inventor
Vivian Hsiun
Alexander G. Macinnis
Xiaodong Xie
Original Assignee
Broadcom Corporation
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Broadcom Corporation filed Critical Broadcom Corporation
Publication of WO2003085494A2 publication Critical patent/WO2003085494A2/fr
Publication of WO2003085494A3 publication Critical patent/WO2003085494A3/fr

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/38—Concurrent instruction execution, e.g. pipeline or look ahead
    • G06F9/3861—Recovery, e.g. branch miss-prediction, exception handling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/38—Concurrent instruction execution, e.g. pipeline or look ahead
    • G06F9/3877—Concurrent instruction execution, e.g. pipeline or look ahead using a slave processor, e.g. coprocessor
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/12—Selection from among a plurality of transforms or standards, e.g. selection between discrete cosine transform [DCT] and sub-band transform or selection between H.263 and H.264
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/12—Selection from among a plurality of transforms or standards, e.g. selection between discrete cosine transform [DCT] and sub-band transform or selection between H.263 and H.264
    • H04N19/122—Selection of transform size, e.g. 8x8 or 2x4x8 DCT; Selection of sub-band transforms of varying structure or type
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/129—Scanning of coding units, e.g. zig-zag scan of transform coefficients or flexible macroblock ordering [FMO]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/102—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or selection affected or controlled by the adaptive coding
    • H04N19/13—Adaptive entropy coding, e.g. adaptive variable length coding [AVLC] or context adaptive binary arithmetic coding [CABAC]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/134—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the element, parameter or criterion affecting or controlling the adaptive coding
    • H04N19/157—Assigned coding mode, i.e. the coding mode being predefined or preselected to be further used for selection of another element or parameter
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
    • H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
    • H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
    • H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/42—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation
    • H04N19/423—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation characterised by memory arrangements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/44—Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/60—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding
    • H04N19/61—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using transform coding in combination with predictive coding
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/70—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals characterised by syntax aspects related to video coding, e.g. related to compression standards
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/80—Details of filtering operations specially adapted for video compression, e.g. for pixel interpolation
    • H04N19/82—Details of filtering operations specially adapted for video compression, e.g. for pixel interpolation involving filtering within a prediction loop
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
    • H04N19/90—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals

Definitions

  • Patent Applications are related to the present application and are hereby specifically incorporated by reference: Patent Application No. 10/114,798, entitled “VIDEO DECODING SYSTEM SUPPORTING MULTIPLE STANDARDS”; Patent Application No. 10/114,679, entitled “METHOD OF OPERATING A VIDEO DECODING SYSTEM”; Patent Application No. 10/114,797, entitled “METHOD OF COMMUNICATING BETWEEN MODULES
  • Patent Application No. 10/114,886 entitled “MEMORY SYSTEM FOR VIDEO DECODING SYSTEM”
  • Patent Application No. 10/114,619 entitled “INVERSE DISCRETE COSINE TRANSFORM SUPPORTING MULTIPLE DECODING PROCESSES”
  • Patent Application No. 10/113,094 entitled “RISC PROCESSOR SUPPORTING ONE OR MORE
  • the present invention relates generally to video decoding systems, and, more particularly, to variable-length decoding.
  • Digital video decoders decode compressed digital data that represent video images in order to reconstruct the video images.
  • Most transmitted video data is compressed and decompressed using, among other techniques, variable-length coding, such as Huffman coding.
  • Huffman coding is a widely used technique for lossless data compression that achieves compact data representation by taking advantage of the statistical characteristics of the source.
  • the Huffman code is a prefix-free variable- length code that assures that a code is uniquely decodable. In Huffman code, no codeword is the prefix of any other codeword.
  • run-length processed data are often subsequently coded by variable-length coding for further data compression.
  • Variable-length encoding following the Huffman coding principle allocates codes of different lengths to different input data according to the probability of occurrence of the input data, so that statistically more frequent input codes are allocated shorter codes than the less frequent codes. The less frequent input codes are allocated longer codes.
  • the allocation of codes may be done either statically or adaptively. For the static case, the same output code is provided for a given input datum, no matter what block of data is being processed.
  • output codes are assigned to input data based on a statistical analysis of a particular input block or set of blocks of data, and possibly changes from block to block (or from a set of blocks to a set of blocks).
  • DSP digital signal processor
  • MAC multiply-accumulate
  • VLIW Very Long Instruction Word
  • the processors are difficult to program when compared to a general-purpose processor, and VLIW processors tend to have difficulty decoding variable length codes since the nature of the codes does not lend itself to parallel operations.
  • special processing accelerators are tightly coupled to the instruction pipeline and are part of the core of the main processor.
  • One aspect of the present invention is directed to a video decoding system comprising a decoder processor, a first variable-length decoding accelerator and a second variable-length decoding accelerator.
  • the decoder processor is adapted to perform decoding functions on a video data stream.
  • the first and second variable- length decoding accelerators are each coupled to the decoder processor and are adapted to perform variable-length decoding operations on variable-length code in the video data stream.
  • Another aspect of the present invention is directed to a variable-length decoder having a plurality of code tables and a code table selection register. The code tables are stored in memory. Each code table corresponds to either a different class of variable length codes in a decoding method or to a different decoding method.
  • Each of the code tables matches variable-length codes to their corresponding decoded information.
  • the code table selection register holds a value that dictates which of the plurality of code tables is to be utilized to decode variable-length code.
  • the register is programmable to dictate the appropriate code table to be employed according to the format of an incoming data stream.
  • Another aspect of the present invention is directed to a video decoding system having a decoder processor and a variable-length decoding accelerator.
  • the decoder processor performs decoding functions on a video data stream.
  • the variable-length decoding accelerator is coupled to the decoder processor and performs variable-length decoding operations on variable-length codes in the video data stream.
  • the variable- length decoding accelerator is capable of decoding variable-length codes according to any of a plurality of decoding methods.
  • FIG. 1 is a functional block diagram of a digital media system in which the present invention may be illustratively employed.
  • FIG. 2 is a functional block diagram demonstrating a video decode data flow according to an illustrative embodiment of the present invention.
  • FIG. 3 is a functional block diagram of a decoding system according to an illustrative embodiment of the present invention.
  • FIG. 4 is a functional block diagram representing a variable-length decoding system according to an illustrative embodiment of the present invention.
  • FIG. 5 is a flowchart representing a method of variable-length decoding a digital video data stream according to an illustrative embodiment of the present invention.
  • FIG. 6 is a block diagram showing stream funnel and codeword search engine elements of a variable-length decoder according to an illustrative embodiment of the present invention.
  • FIG. 7 is a flowchart representing a method of decoding a variable-length code data stream according to an illustrative embodiment of the present invention.
  • FIG. 8 is an example of a code table according to the code table storage algorithm of the present invention.
  • FIG. 9 is a chart representing a decoding pipeline according to an illustrative embodiment of the present invention.
  • FIG. 10 is a flowchart representing a macroblock decoding loop according to an illustrative embodiment of the present invention.
  • FIG. 11 is a functional block diagram of a digital video decoding system according to an illustrative embodiment of the present invention.
  • FIG. 12 is a chart representing a decoding pipeline according to an illustrative embodiment of the present invention.
  • FIG. 13 is a chart representing a dual-row decoding pipeline employing cycle stealing according to an illustrative embodiment of the present invention.
  • FIG. 1 is a functional block diagram of a digital media system in which the present invention may be illustratively employed. It will be noted, however, that the present invention can be employed in systems of widely varying architectures and widely varying designs.
  • the digital media system of FIG. 1 includes transport processor 102, audio decoder 104, direct memory access (DMA) controller 106, system memory controller
  • the transport processor 102 receives and processes a digital media data stream.
  • the transport processor 102 provides the audio portion of the data stream to the audio decoder 104 and provides the video portion of the data stream to the digital video decoder 116.
  • the audio and video data is stored in main memory 110 prior to being provided to the audio decoder 104 and the digital video decoder 116.
  • the audio decoder 104 receives the audio data stream and produces a decoded audio signal.
  • DMA controller 106 controls data transfer amongst main memory 110 and memory units contained in elements such as the audio decoder 104 and the digital video decoder 116.
  • the system memory controller 108 controls data transfer to and from system memory 110.
  • system memory 110 is a dynamic random access memory (DRAM) unit.
  • the digital video decoder 116 receives the video data stream, decodes the video data and provides the decoded data to the display engine 120 via the display feeder 118.
  • the analog video decoder 126 digitizes and decodes an analog video signal (e.g. NTSC or PAL) and provides the decoded data to the display engine 120.
  • the graphics engine 122 processes graphics data in the data stream and provides the processed graphics data to the display engine 120.
  • the display engine 120 processes graphics data in the data stream and provides the processed graphics data to the display engine 120.
  • FIG. 2 is a functional block diagram demonstrating a video decode data flow according to an illustrative embodiment of the present invention.
  • Transport streams are parsed by the transport processor 102 and written to main memory 110 along with access index tables.
  • the video decoder 116 retrieves the compressed video data for decoding, and the resulting decoded frames are written back to main memory 110.
  • Decoded frames are accessed by the display feeder interface 118 of the video decoder for proper display by a display unit.
  • two video streams are shown flowing to the display engine 120, suggesting that, in an illustrative embodiment, the architecture allows multiple display streams by means of multiple display feeders.
  • aspects of the present invention relate to the architecture of digital video decoder 116.
  • a moderately capable general purpose CPU with widely available development tools is used to decode a variety of coded streams using hardware accelerators designed as integral parts of the decoding process.
  • FIG. 3 is a functional block diagram of a digital video decoding system 300 according to an illustrative embodiment of the present invention.
  • the digital video decoding system 300 of FIG. 3 can illustratively be employed to implement the digital video decoder 116 of Figures 1 and 2.
  • Video decoding system 300 includes core decoder processor 302, DMA Bridge 304, decoder memory 316, display feeder 318, phase-locked loop element 320, and acceleration modules 306, 308, 309, 310, 312, 313 and 315.
  • the acceleration modules include variable-length decoder (VLD) 306, inverse quantization (IQ) module 308, inverse discrete cosine transform (IDCT) module 309, pixel filter 310, motion compensation module 312, loop filter 313 and post filter 315.
  • VLD variable-length decoder
  • IQ inverse quantization
  • IDCT inverse discrete cosine transform
  • the acceleration modules 306, 308, 309, 310 and 312 are hardware accelerators that accelerate special decoding tasks that
  • the core processor 302 is the central control unit of the decoding system 300.
  • the core processor 302 prepares the data for decoding.
  • the core processor 302 also orchestrates the macroblock (MB) processing pipeline for the acceleration modules and fetches the required data from main memory 110 via the DMA bridge 304.
  • the core processor 302 also handles some data processing tasks.
  • Picture level processing including sequence headers, GOP headers, picture headers, time stamps, macroblock- level information, except the block coefficients, and buffer management, are performed directly and sequentially by the core processor 302, without using the accelerators 304, 306, 308, 309, 310, 312, 313 and 315, other than the VLD 306 (which accelerates general bitstream parsing).
  • Picture level processing does not generally overlap with slice level/macroblock decoding.
  • the core processor 302 is a MIPS processor, such as a MIPS32 implementation, for example.
  • variable-length decoder 306 implements some or all of the functions implemented by variable-length decoder 306, and the other hardware accelerators 308, 309, 310 312, 313 and 315, in different ways that prevent fixed hardware implementations from addressing all requirements without duplication of resources.
  • variable-length decoder 306 and the other hardware accelerators 308, 309, 310, 312, 313 and 315 are internally programmable to allow changes according to various processing algorithms. This enables a decoding system that decodes most standards efficiently and flexibly.
  • the decoding system of the present invention employs high-level granularity acceleration with internal programmability to achieve the requirements above by implementation of very fundamental processing structures that can be configured dynamically by the core decoder processor.
  • MAC multiply-accumulate
  • the decompression algorithm has to be implemented with firmware that uses individual low-level instructions (like MAC) to implement a high-level function, and each instruction runs on the core processor.
  • the firmware configures, i.e., programs, variable-length decoder 306 and the other hardware accelerators 308, 309, 310, 312, 313 and 315, which in turn represent high-level functions (like variable-length decoding) that run without intervention from the main core processor 302. Therefore, each hardware accelerator 306, 308, 309, 310, 312, 313 and 315, runs in parallel according to a processing pipeline dictated by the firmware in the core processor 302. Upon completion of the high-level functions, each accelerator notifies the main core processor 302, which in turn decides what the next processing pipeline step should be.
  • the software control consists of a simple pipeline that orchestrates decoding by issuing commands to each hardware accelerator module for each pipeline stage, and a status request mechanism that makes sure that all modules have completed their pipeline tasks before issuing the start of the next pipeline stage.
  • stage can refer to all of the decoding functions performed during a given time slot, or it can refer to a functional step, or group of functional steps, in the decoding process.
  • Each hardware module 306, 308, 309, 310, 312, 313 and 315 performs its task after being so instructed by the core processor 302.
  • each hardware module includes a status register that indicates whether the module is active or inactive.
  • the core processor 302 polls the status register to determine whether the hardware module has completed its task.
  • the hardware accelerators share a status register.
  • Variable-length decoder 306 is a hardware accelerator that accelerates the process of decoding variable-length codes, which might otherwise be a bottleneck for a decoding process if it were handled by the core processor 302 alone.
  • the VLD 306 can also be any other type of entropy decoder. But for purposes of explanation, the present invention will be described with respect to a variable-length decoder.
  • the VLD 306 performs decoding of variable length codes (VLC) in the compressed bit stream to extract coefficients, such as DCT coefficients, from the compressed data stream.
  • VLC variable length codes
  • Different coding formats generally have their own special VLC tables.
  • the VLD module 306 is internally programmable to allow changes according to various processing algorithms.
  • the VLD 306 is completely configurable in terms of the VLC tables it can process.
  • the VLD 306 can accommodate different VLC tables, selectable as needed under the control of the core processor 302.
  • the VLD 306 includes a register that the core processor can program to guide the VLD 306 to use the appropriate VLC table according to the needs of the encoding/decoding algorithm and the class of codes expected.
  • the VLD 306 is designed to support the worst-case requirement for VLD operation, such as with MPEG-2 HDTV (Main Profile at High Level) for video decoding, while retaining its full programmability.
  • the VLD 306 includes a code table random access memory (RAM) for fastest performance.
  • RAM code table random access memory
  • Some compression/decompression formats such as Windows Media Technology 8 (WMT8) video, may require larger code tables that do not fit entirely within the code RAM in the VLD 306.
  • WMT8 Windows Media Technology 8
  • the VLD 306 can make use of both the decoder memory 316 and the main memory 110 as needed. Performance of VLC decoding is reduced somewhat when codes are searched in video memory 316 and main memory 110. Therefore, for formats that require large amounts of code, the most common codes are stored in the
  • VLD code RAM the next most common codes are stored in decoder memory 316, and the least common codes are stored in main memory 110. Also, such codes are stored in decoder memory 316 and main memory 110 such that even when extended look-ups in decoder memory 316 and main memory 110 are required, the most commonly occurring codes are found more quickly. This allows the overall performance to remain exceptionally high.
  • the VLD 306 decodes variable length codes in as little as one clock, depending on the specific code table in use and the specific code being decoded.
  • the VLD 306 helps the core processor 104 to decode header information in the compressed bitstream.
  • the VLD module 306 is architected as a coprocessor to the decoder processor 110. That is, it can operate on a single- command basis where the core processor issues a command (via a coprocessor instruction) and waits (via a Move From Coprocessor instruction) until it is executed by the VLD 306, without polling to determine completion of the command. This increases performance when a large number of VLC codes that are not DCT coefficients are parsed.
  • the VLD 306 is architected as a hardware accelerator.
  • the VLD 306 can perform complex tasks such as decoding a set of VLC codes, and it includes a status register that indicates whether the module is active or inactive.
  • the core processor 302 polls the status register to determine whether the VLD 306 has completed its tasks.
  • the VLD 306 shares a status register with other decoding elements, such as decoding elements 308, 309, 310 and 312.
  • the VLD module 306 includes two variable-length decoders.
  • Each of the two variable-length decoders can be hardwired to efficiently perform decoding according to a particular compression standard, such as MPEG-2 HD.
  • one or both of two VLDs can be optionally set as a programmable VLD engine, with a code RAM to hold VLC tables for other media coding formats.
  • the two VLD engines are controlled independently by the core processor 302, and either one or both of them will be employed at any given time, depending on the application.
  • the VLD 306 can operate on a block-command basis where the core processor 302 commands the VLD 306 to decode a complete block of VLC codes, such as DCT coefficients, and the core processor 302 continues to perform other tasks in parallel. In this case, the core processor 302 verifies the completion of the block operation by checking a status bit in the VLD 306. The VLD 306 produces results
  • tokens that are stored in decoder memory 316.
  • the VLD 306 checks for invalid codes and recovers gracefully from them. Invalid codes may occur in the coded bit stream for a variety of reasons, including errors in the video encoding, errors in transmission, and discontinuities in the stream.
  • the inverse quantizer module 308 performs run-level code (RLC) decoding, inverse scanning (also called zig-zag scanning), inverse quantization and mismatch control.
  • the coefficients, such as DCT coefficients, extracted by the VLD 306 are processed by the inverse quantizer 308 to bring the coefficients from the quantized domain to the DCT domain.
  • the IQ module 308 obtains its input data (run-level values) from the decoder memory
  • the IQ module 308 obtains its input data directly from the VLD 306.
  • This alternative embodiment is illustratively employed in conjunction with encoding/decoding algorithms that are relatively more involved, such as MPEG-2 HD decoding, for best performance.
  • the run-length, value and end-of-block codes read by the IQ module 308 are compatible with the format created by the VLD module when it decodes blocks of coefficient VLCs, and this format is not dependent on the specific video coding format being decoded.
  • the IDCT module 309 performs the inverse transform to convert the coefficients produced by the IQ module 308 from the frequency domain to the spatial domain.
  • the primary transform supported is the discrete cosine transform (DCT) as specified in MPEG-2, MPEG-4, IEEE, and several other standards.
  • the IDCT module 309 also supports alternative related transforms, such as the "linear" transform in H.264 and MPEG-4 AVC, which is not quite the same as IDCT.
  • the coefficient input to the IDCT module 309 is read from decoder memory 316, where it was placed after inverse quantization by the IQ module 308. The transform result is written back to decoder memory 316.
  • the IDCT module uses the same memory location in decoder memory 316 for both its input and output, allowing a savings in on-chip memory usage.
  • the coefficients produced by the IQ module are provided directly to the IDCT module 309, without first depositing them in decoder memory 316.
  • the IQ module 308 and IDCT module 309 are part of the same hardware module and use a common interface to the core processor.
  • the transfer of coefficients from the IQ module 308 to the IDCT module 309 can be either direct or via decoder memory 316. For encoding/decoding algorithms that are relatively more involved, such as MPEG-2 HD decoding, the transfer is direct in order to save time and improve performance.
  • the pixel filter 310 performs pixel filtering and interpolation as part of the motion compensation process.
  • Motion compensation is performed when an image from a region of a previous frame is similar to a region in the present frame, just at a different location within the frame. Rather than recreate the image anew from scratch, the previous image is used and just moved to the proper location within the frame. For example, assume the image of a person's eye is contained in a macroblock of data at frame #0. Say that the person moved to the right so that at frame #1 the same eye is located in a different location in the frame. Motion compensation uses the eye from frame #0 (the reference frame) and simply moves it to the new location in order to get the new image.
  • the new location is indicated by motion vectors that denote the spatial displacement in frame #1 with respect to reference frame #0.
  • the pixel filter 310 performs the interpolation necessary when a reference block is translated (motion-compensated) into a position that does not land on whole- pixel locations.
  • a hypothetical motion vector may indicate to move a particular block 10.5 pixels to the right and 20.25 pixels down for the motion- compensated prediction.
  • the motion vectors are decoded by the VLD 306 in a previous processing pipeline stage and are stored in the core processor 302.
  • the pixel filter 310 gets the motion infonnation as vectors and not just bits from the bitstream during decoding of the "current" macroblock in the "current" pipeline stage.
  • the reference block data for a given macroblock is stored in memory after decoding of said macroblock is complete.
  • the reference picture data is stored in system memory 110.
  • the pixel filter 310 retrieves the reference macroblock pixel information from system memory 110 and the motion vector from the core processor 302 and performs pixel filtering.
  • the pixel filter stores the filtered result (pixel prediction data) in decoder memory 316.
  • the motion compensation module 312 reconstructs the macroblock being decoded by performing the addition of the decoded difference (or "error") pixel information from the IDCT 309 to the pixel prediction data from the output of the pixel filter 310.
  • the pixel filter 310 and motion compensation module 312 are shown as one module in FIG. 3 to emphasize a certain degree of direct cooperation between them.
  • the loop filter 313 and post filter 315 perform de-blocking filter operations. Some decoding algorithms employ a loop filter and others employ a post filter. The difference is where in the processing pipeline each filter 313, 315 does its work.
  • the loop filter 313 processes data within the reconstruction loop and the results of the filter are used in the actual reconstruction of the data.
  • the post filter 315 processes data that has already been reconstructed and is fully decoded in the two-dimensional picture domain.
  • the loop filter 313 and post filter 315 are combined in one filter module.
  • the input data to the loop filter 313 and post filter 315 comes from decoder memory 316.
  • This data includes pixel and block/macroblock parameter data generated by other modules in the decoding system 300.
  • the loop filter 313 and post filter 315 have no direct interfaces to other processing modules in the decoding system 300.
  • the output data from the loop filter 313 and post filter 315 is written into decoder memory 316.
  • the core processor 302 then causes the processed data to be put in its correct location in main memory.
  • the core processor 302 interprets the decoded bits for the appropriate headers and decides and coordinates the actions of the hardware blocks 306, 308, 309, 310, 312, 313 and 315. Specifically, all macroblock header information, from the macroblock address increment (MBAinc) to motion vectors (MVs) and to the cbp pattern, in the case of MPEG-2 decoding, for example, is derived by the core processor 302. The core processor 302 stores related information in a particular format or data structure (determined by the hardware module specifications) in the appropriate buffers in the decoder memory 316.
  • the quantization scale is passed to the buffer for the IQ engine 308; macroblock type, motion type and pixel precision are stored in the parameter buffer for the pixel filter engine 310.
  • the core processor keeps track of certain infonnation in order to maintain the coreect pipeline. For example, motion vectors of the macroblock are kept as the predictors for future motion vector derivation.
  • Decoder memory 316 is used to store macroblock data and other time-critical data used during the decode process.
  • decoder memory 316 accesses decoder memory 316 to either read the data to be processed or write processed data back.
  • all currently used data is stored in decoder memory 316 to minimize access to main memory.
  • Each hardware module 306, 308, 309, 310, 312, 314 is assigned one or more buffers in decoder memory 316 for data processing. Each module accesses the data in decoder memory 316 as the macroblocks are processed through the system.
  • decoder memory 316 also includes parameter buffers that are adapted to hold parameters that are needed by the hardware modules to do their job at a later macroblock pipeline stage. The buffer addresses are passed to the hardware modules by the core processor 302.
  • decoder memory 316 is a static random access memory (SRAM) unit.
  • the core processor 302, DMA Bridge 304, VLD 306, IQ 308, IDCT 309, pixel filter 310, motion compensation module 312, loop filter 313 and post filter 315 have access to decoder memory 316 via the internal bus 322.
  • the decoder 315 use the decoder memory 316 as the source and destination memory for their nonnal operation.
  • the CPU 114 has access to decoder memory 316, and the DMA engine 304 can transfer data between decoder memory 316 and the main system memory (DRAM) 110.
  • the arbiter for decoder memory 316 is in the bridge module 304.
  • the bridge module 304 arbitrates and moves picture data between decoder memory 316 and main memory.
  • the bridge interface 304 includes an internal bus network that includes arbiters and a direct memory access (DMA) engine.
  • the DMA bridge 304 serves as an asynchronous interface to the system buses.
  • the display feeder module 318 reads decoded frames from main memory and manages the horizontal scaling and displaying of picture data.
  • the display feeder 318 interfaces directly to a display module.
  • the display feeder 318 includes multiple feeder interfaces, each including its own independent color space converter and sealer.
  • the display feeder 318 handles its own memory requests via the bridge module 304.
  • the core processor 302 runs at twice the frequency of the other processing modules 306, 308, 309, 310, 312, 313, 315.
  • the system clock signal CLKLN is used as input to the phase-locked loop element (PLL) 320, which is a closed-loop feedback control system that locks to a particular phase of the system clock to produce a stable signal with little jitter.
  • the PLL element 320 generates a IX clock for the hardware accelerators, DMA bridge 304 and the core processor bus interface, while generating a 2X clock for the core processor 302 and the core processor bus interface. This is to allow the core processor 302 to operate at high clock frequencies if it is designed to do so, and to allow other logic to operate at the slower lx clock frequency. It also allows the decoding system 300 to run faster than the nominal clock frequency if the circuit timing supports it.
  • FIG. 4 is a functional block diagram representing a variable-length decoding system according to an illustrative embodiment of the present invention.
  • elements that are also shown in FIG. 3 are given like reference numbers.
  • the VLD variable-length decoding system
  • 306 includes decoder processor interface 400, stream funnel 402, codeword search engine 404, block buffer 406, decoder memory interface 408, code table selection register 412 and status register 414.
  • the input 410 to the VLD 306 is a bit stream without explicit word boundaries.
  • the VLD 306 decodes a codeword, determines its length, and shifts the input data stream by the number of bits conesponding to the decoded code length, before decoding the next codeword. These are recursive operations that are not pipelined.
  • the VLD 306 is implemented based on a small, local, code table memory unit, located in codeword search engine 404, that stores programmable variable length code tables.
  • the local memory unit is a random access memory (RAM) unit.
  • a small code table memory unit is achieved by employing a multistage search structure that reduces the storage requirement, enables fast bit extraction and efficiently handles the case of a large number of code tables.
  • the stream funnel 402 receives data from the source (or coded data buffer) and shifts the data according to the previously decoded code length, so as to output the correct window of bits for the symbols that are being currently decoded.
  • the stream funnel receives the incoming bitstream 410 from system memory 110.
  • the codeword search engine 404 mainly behaves as a symbol search engine.
  • the codeword search engine is based on a multistage search structure. Since codewords are usually assigned based on the probability of appearance, the shortest codeword is generally assigned to the most frequent appearance.
  • the multistage search structure is based on this concept.
  • the codeword search engine 404 incorporates a small code memory that is employed for performing pattern matching.
  • a multistage, pipelined structure is employed to handle the case of a long codeword.
  • a code table reduction algorithm can further reduce the storage requirement for a large number of code tables.
  • Status register 414 is adapted to hold an indicator of the status of the VLD
  • the status register is accessible by the core decoder processor 302 to determine the status of VLD 306.
  • the status register 414 indicates whether or not the VLD has completed its variable-length decoding functions on the current macroblock.
  • the VLD module 306 is architected as a coprocessor to the decoder processor 302. That is, the VLD 306 can operate on a single-command basis where the core processor issues a command (via a coprocessor instruction) and waits (via a Move From Coprocessor instruction) until it is executed by the VLD 306, without polling the status register 414 to determine completion of the command.
  • Code table selection register 412 is adapted to hold a value that dictates which of a plurality of VLD code tables is to be utilized to decode variable-length code.
  • code table selection register 412 holds the starting address of the code table to be employed.
  • the code table selection register 412 is programmable to dictate the appropriate code table to be employed according to the format of an incoming data stream and the class of variable length codes that are expected next.
  • the core video processor 302 provides a value (an address, for example) to register 412 to point to the code table that is appropriate for the current data stream and the state of decoding the current stream.
  • the code tables can be switched on a syntax element basis, a macroblock-to- macroblock basis or more or less frequently, as required by the application.
  • FIG. 5 is a flowchart representing a method of variable-length decoding a digital video data stream according to an illustrative embodiment of the present invention.
  • video data of a first encoding/decoding format is received.
  • variable-length decoder 306 is configured based on the first encoding/decoding format.
  • variable-length decoder 306 configures variable-length decoder 306 by programming code table selection register 412.
  • video data of the first encoding/decoding format is decoded by the variable-length decoder 306.
  • video data of a second encoding/decoding format is received.
  • the variable-length decoder 306 is configured based on the second encoding/decoding fonnat.
  • video data of the second encoding/decoding fonnat is decoded using the variable-length decoder 306.
  • the programming for different decoding fonnats is done through register bus read and write.
  • FIG. 6 is a block diagram showing the stream funnel 402 and codeword search engine 404 elements of VLD 306, according to an illustrative embodiment of the present invention.
  • Stream funnel 402 includes data stream input buffer 600, register Do 602, register Di 604, left-shifter 606, register D 608, and accumulator 610.
  • the input data (coded stream) are stored in input buffer 600, which, in an illustrative embodiment, is a first-in first-out (FIFO) buffer.
  • the input buffer 600 provides the data to register Do 602.
  • Register Di 604 also stores part of the incoming bitstream by virtue of load operations that will be discussed below and which load data from register Do 602 into register Di.
  • the contents of registers Do and Di are in turn provided to left shifter 606.
  • registers Do and Di comprise a number of bits equal to the maximum code length.
  • registers Do 602 and Di 604 each are 32-bit registers, and left-shifter 606 can hold up to 64 bits.
  • Register D 2 608 indicates the number of bits in register Di 604 for which the codeword search engine 604 most recently performed a codeword search. If registers Do and Di each hold 32 bits, the number of bits indicated by register D 2 can lie between 0 and 31. This number controls the left shifter 606. After the codeword search engine 404 performs a codeword search for a group of bits in register Dj, register D 2 indicates the number of bits just searched.
  • Left shifter 606 then shifts the indicated number of bits to the left so that the first un- searched bit appears at the most significant bit of the output of the left shifter 606.
  • Accumulator 610 accumulates the number of bits in register Di 604 that have been searched by codeword search engine 404 over multiple codeword searches. When the accumulated code length (the number of bits that have been searched) is greater than or equal to the size of register Di 604 (for example, 32 bits), a carry-out bit 612 becomes 1. This indicates that all the bits in register Di 604 have been used and that register D 0 might not contain the whole next codeword. In that case, a "load" signal is generated.
  • register Do 602 When the "load” signal is generated, the contents of register Do 602 are loaded into register Di 604, a new data word (32 bits in the illustrative example) from the input buffer 600 is loaded into D 0 , and the left shifter 606 shifts by the number of bits indicated by register D 608 to the new position, all at substantially the same time, to prepare for the next search/decode cycle. If the accumulated code length is not greater than or equal to the size of register Di 604 (e.g., 32), the carry- out signal 612 is 0.
  • registers Do 602 and Di 604 Assuming the maximum code length is 32 bits (the size of registers Do 602 and Di 604 in the illustrative embodiment), since at least 32 bits of data in register Do 602 and Di 604 are not used yet, there are always enough bits for the next search/decoding cycle. Registers Do 602 and Di 604 remained unchanged if the accumulated code length is not greater, than or equal to the size of the registers Do
  • the decoding pipes When the accumulated code length is greater than or equal to the size of registers Do and Di, and there is no data available in the input buffer 600, the decoding pipes are put on hold. In other words, the contents of register Do 602 are not loaded into register Di 604. The decoding processing then waits until data is available in the input buffer 600.
  • Codeword search engine 404 includes an address generator 612 and a local memory unit 514.
  • Address generator 612 generates a memory address at which to perform a codeword search. In an illustrative embodiment, this address will reside in the local memory unit 614, but it may also reside in decoder memory 316 or system memory 110, as will be described below.
  • the address generator 612 generates the address to be searched by adding the value of the bits retrieved from left shifter 606, i.e., the data for which a search is to be perfonned, to a base address. For the first search performed in a given code table, and for subsequent searches when the previous search yielded a code match, the base address is equal to the start address of the code table to be searched.
  • the base address is equal to the sum of the start address of the code table plus an offset that was indicated by the code table entry of the previous search.
  • the starting address of the code table to be searched can be programmed. In this way, the appropriate code table can be selected for the encoding/decoding fonnat and the current state of the bitstream being decoded.
  • code table selection register 412 holds the starting address of the code table to be searched. This register can be accessed by the decoder processor 302 to point to the code table that is appropriate for the current data stream.
  • the code tables can be switched on a syntax element basis, a macroblock-to-macroblock basis or on other intervals.
  • Local code table memory 614 holds the code look-up table that is to be used during the variable-length decoding process.
  • the code table that starts at the indicated start address is used in decoding the incoming bitstream.
  • code table memory 614 is a random access memory (RAM) unit.
  • the code table memory 614 is a relatively small memory unit, for example, a 512x32 single-port RAM.
  • portions of the table can be stored in decoder memory 316 and/or system memory 110.
  • decoder memory 316 if more memory is needed than the local memory unit 614 alone, first the decoder memory 316 is utilized, and if more still is needed, the system memory 110 is utilized. Where multiple memory units are utilized, the shortest, and therefore most common codes, are stored in local code table memory 614. The next-shortest codes are stored in decoder memory 316, and if needed, the longest codes are stored in system memory 110. This architecture allows for fast bit extraction.
  • the codeword search engine 404 employs a code table storage and look-up method that enables fast bit extraction and also reduces the size of the code tables. Reducing the size of the code tables further reduces the storage requirement for a large number of code tables and results in fast performance for a broad range of codes.
  • One embodiment of the code table storage and look-up method makes use of the multiple memory unit structure mentioned above and uses a multistage, pipelined structure to handle the case of a long codeword.
  • the code table memory unit 614 supports multiple code tables (up to 32 in an illustrative embodiment).
  • each code table has the following general information which is pre-programmed by the decoder processor 302: the starting address, in the local memory 614, of the code table during the first search (FSA), the searching length during the first level search (FSL), an indication of whether a sign bit follows the code to be searched, the size of a run code that may follow a specific variable length code (such as an escape code), the size of a level code that may follow a specific variable length code (such as an escape code), and an indication of whether an end of block (EOB) code is to follow the run-level code that may appear.
  • a high sign bit indicator indicates that the code table has a sign bit following the codeword.
  • the size of the run code indicates how many bits are allocated to the run portion of the results.
  • the size of the level code indicates how many bits are allocated to the level portion of the results.
  • the EOB bit indicates whether a "last" bit or EOB bit is expected after the run-level code that may appear.
  • the run-level code and EOB bit may appear following a designated variable length code such as escape (ESC) code.
  • ESC escape
  • the 15 bits following ESC are decoded as fixed length codes represented by 1- bit LAST, 6-bit RUN and 8-bit LEVEL.
  • the meanings of run, level and EOB can vary between different video coding/decoding fonnats.
  • Each address of a code table comprises a code table entry.
  • Each entry includes a current code length (CCL) indicator, a next search length (NSL) indicator, an end-of-block (EOB or last code) bit, a status indicator and an information/offset value.
  • the status indicator indicates whether that entry represents a codeword match. If the entry does represent a codeword match, the information offset value is the matching information, that is, the data that the just-matched codeword represents (the "meaning" of the codeword). If the entry does not represent a codeword match, the infonnation offset value indicates an address at which to perform the next codeword search. In an illustrative embodiment of the present invention, the offset value indicates an address at which to base the next search to complete decoding of the current codeword.
  • the offset value is added to another address to obtain the base address from which to perforai the next search to complete decoding of the cunent codeword.
  • the status indicator can also indicate other aspects of the search status. For example, if the entry does not represent a codeword match, the status indicator indicates the memory unit in which to perform the next codeword search. Also, if the entry represents an enor, i.e., no valid code would result in the entry at that memory location to be reached, the status indicator indicates as much. In an illustrative embodiment of the present invention, the status indicator is a 4-bit word having the meanings shown in Table 1.
  • the information/offset value represents the offset of the tertiary code table at the system memory 110.
  • the current code-length indicator indicates the number of bits that the input bitstream should be shifted prior to the next codeword search. If the code table entry represents a codeword match, the current code-length represents the number of bits, out of the currently searched group of bits, that correspond to the matched information represented by the information/offset value. If the code table entry does not represent a codeword match, the current code-length indicator indicates the number of bits that were consumed in the cunent stage of search.
  • the cunent code-length indicator indicates that no bits in the current search have been matched.
  • the cunent code-length indicator indicates the number of bits consumed from the input bit stream when the status indicates a match, and it indicates the number of bits to be searched in the next stage of search when the status indicates no match in the current stage. In such an embodiment the number of bits consumed when there is not a match is implied to be the number of bits searched in the current stage.
  • each code table entry that does not represent a codeword match further includes a next-search-length (NSL) indicator that indicates the number of bits to perform a codeword search for in the next stage.
  • NSL next-search-length
  • the code table entries that do represent a codeword match do not contain a next-search-length indicator, as the search length in the next stage automatically reverts to an initial value.
  • the code table entries that do represent a codeword match do contain a next-search-length indicator, which indicates the initial value.
  • the end-of-block bit is high if the just-decoded code is the last code in a block of codes to be decoded. In alternative embodiments the end-of-block bit has the opposite polarity, or the end-of-block bit is not included.
  • the code table memory 614 and the address generator 612 work together to perfonn pattern matching on the data stream. When a codeword is matched at a code table entry, the status indicator in the entry will indicate that that is the case. If an accessed code table entry is not a match, the state machine will go to the next stage to keep searching until the codeword is found. If the status indicator shows that an enor has occurred, the VLD 306 will stop searching the next codeword, set an enor status bit to "1 ,” report the enor to the decoder processor 302 and enter an idle state.
  • FIG. 7 is a flowchart representing a method of decoding a variable-length code data stream according to an illustrative embodiment of the present invention.
  • the method implements a code table storage algorithm, and a method of traversing a code table implementing the algorithm, that reduces the storage requirement and enables fast code look-up.
  • the appropriate code table is loaded according to the compression/decompression standard of the data stream being decoded.
  • the code table is illustratively loaded into local memory 614.
  • the start address of the code table in the local memory is designated m.
  • a base memory address is set equal to the start address.
  • the search length, n i.e., the number of bits from the data stream for which a code match is sought in a given search, is initialized as a first search length (FSL) value.
  • FSL search length
  • the next n bits in the data stream are retrieved. In an illustrative embodiment, these bits are retrieved from the n most significant bits of left shifter 606.
  • the address at which to search for a code match is generated by adding the value of the n bits retrieved from the bitstream to the base address. This step is illustratively performed by address generator 612.
  • the memory location having the address generated in step 720 is accessed, and the status indicator at that memory location is examined. Decision box 730 asks whether the status indicator indicates that a codeword match is found. If the answer is yes, the conesponding infonnation (decoded data), indicated by the information/offset value of the memory location, is output.
  • the decoded data comprises run and level information which implies the value of one or more transfonn coefficients, such as discrete cosine transform (DCT) coefficients.
  • the decoded data comprises transform coefficients directly, or other information as appropriate to the video compression/decompression standard being decoded. If the status indicator indicates that a codeword match is not found, decision box 740 asks whether the status indicator indicates that an error has occurred. Such an enor would arise, for example, if the memory location arrived at does not conespond to a valid code. If there is an enor, an enor indication is given, as indicated at step 745.
  • the base address is set equal to the start address, as indicated by step 755, and the search length, n, is set equal to the first search length (FSL), as shown at step 760. If the status indicator indicates that the memory location does not represent a codeword match, and an error has not occuned, the base address is set according to the offset value indicated by the infonnation/offset value, as indicated at step 750, and the search length, n, is set equal to the next-search-length value held in the memory location. In an illustrative embodiment, the search length remains constant throughout the decoding process. In that case, steps 760 and 765 of FIG. 5 can be eliminated.
  • Step 770 the incoming bitstream is shifted by an amount indicated by the cunent code-length indicator of the memory location.
  • Step 770 is illustratively perfonned by left shifter 606.
  • the current code-length indicator indicates the number of the input bits consumed by the most recent search stage in decoding the cunent code word..
  • the value of the cunent code-length indicator is equal to n bits (the number of bits for which the current search was performed).
  • the status indicator indicates an enor
  • the value of the cunent code-length indicator is zero.
  • step 770 the next n bits in the data stream are accessed, as indicated by step 715, and the above-described process is repeated starting at that point. In an exemplary embodiment, this process is iteratively repeated as long as there is data in the data stream to decode.
  • FIG. 8 is an example of a code table according to the code table storage algorithm of the present invention.
  • the code table of FIG. 8 is stored in local memory 614.
  • the following codebook (Table 2) is used in the exemplary code table of FIG. 8: Codeword Code Length Decoded
  • Each of the addresses 800 in the code table of FIG. 8 represents a codebook entry.
  • the other columns 810, 820, 830, 840 and 850 represent elements of each codebook entry. These elements include current code-length indicator 810, next- search-length indicator 820, end-of-block bit 830, status indicator 840 and information/offset value 850.
  • the illustrative code table of FIG. 8 has a first search length (FSL) of 3 and a starting address (FSA) of 0.
  • the bits in the most-significant position of left shifter 606 are the bits 1010 (which we know, from referring to the codebook of Table 2, represent symbols A and B).
  • the codeword search engine decodes these bits as follows. Because the first search length is 3, the first three bits of the data stream (101) are pulled from the data stream, that is, from the left shifter 606.
  • the address generator 612 adds the value of these bits (5) to the starting address (0) to get a search address of 5.
  • the information offset value (A) of the entry is outputted as a decoded value.
  • the decoded data comprises transform coefficients, such as discrete cosine transform (DCT) coefficients.
  • this decoded value is provided to decoder memory 316 and stored there.
  • the current-code-length indicator of the entry at address 5 is a 1. This value is provided to accumulator 610 and register D 608, indicating that one bit (the first 1, corresponding to the outputted A) was decoded in this stage.
  • the left shifter 606 shifts its contents one bit, putting the bits 010 at the three most-significant positions of left shifter 606.
  • the search length is three (the first search length) because the previous search resulted in a codeword match.
  • the bits 010 are provided to the address generator 612, which adds the value of these bits (2) to the starting address (0) to get a search address of 2 (the starting address is used as the base address because the previous search yielded a match).
  • the input string 1010 has been decoded as AB.
  • the cunent- code-length indicator of the entry at address 2 is a 3. This value is provided to accumulator 610 and register D 2 608, indicating that three bits (010, corresponding to the outputted B) were decoded in this stage.
  • the next bits in the data stream (after the bits 1010) are 00010010 (which represent symbols X and B). Because the value stored in register D 2 608 from the previous search is 3, prior to performing the next search, the left shifter 606 shifts its contents three bits, putting the bits 000 at the three most- significant positions of left shifter 606.
  • the search length is three (the first search length) because the previous search resulted in a codeword match.
  • the bits 000 are provided to the address generator 612, which adds the value of these bits (0) to the starting address (0) to get a search address of 0 (the starting address is used as the base address because the previous search yielded a match).
  • the code table entry at address 0 has a next search-length indicator of 3. This value is provided to address generator 612 to indicate the number of bits to be retrieved from the left shifter 606 for the next search.
  • the current-code-length indicator of the entry at address 0 is a 3. This value is provided to accumulator 310 and register D 608, indicating that the left shifter 606 should shift its contents three bits prior to the next codeword search.
  • the search length is three, as indicated to the address generator 612 by the next-search-length indicator from the previous stage.
  • the bits 100 are provided to the address generator 612, which adds the value of these bits (4) to the base address to get the search address.
  • the base address is equal to the start address (0) plus the offset value (8) indicated by the information/offset value from the previous stage.
  • the search address 0 + 8
  • the current-code-length indicator of the entry at address 12 is a 2. This value is provided to accumulator 610 and register D 2 608, indicating that two bits (10, which are the first two bits of the just- searched bits and which are also the last two bits of the just-decoded codeword) were decoded in this stage.
  • the left shifter 606 shifts its contents two bits, putting the bits 010 at the three most-significant positions of left shifter 606.
  • the search length is three (the first search length) because the previous search resulted in a codeword match.
  • the bits 010 are provided to the address generator 612, which adds the value of these bits (2) to the starting address (0) to get a search address of 2 (the starting address is used as the base address because the previous search yielded a match).
  • the symbol B is decoded at the code table entry at address 2, as was described above.
  • multiple memory units are used to store the codeword look-up table.
  • part of the codeword look-up table is stored in local memory 614, part is stored in decoder memory 316, and part is stored in system memory 110.
  • the shortest, and therefore most common, codes are stored in local memory 614, enabling the majority of codeword searches to be performed quickly and efficiently.
  • the next shortest codes are stored in decoder memory 316 and the longest codes are stored in system memory 110.
  • the status indicator of each code table entry indicates the memory unit at which to perform the next search if the current search did not result in a codeword match. If the current search did produce a codeword match, the status indicator indicates that condition and the next search will be performed in local memory unit 614.
  • the first search for a data stream, and each search following a codeword match are performed in the local memory unit 614.
  • the VLD 306 will continue decoding the bitstream as long as there is space available in the block buffer 406.
  • the VLD 306 In order to simplify the design, in an illustrative embodiment of the present invention, the VLD
  • VLD 306 checks the buffer availability before starting to decode a block. When the VLD 306 is finished decoding a block, the VLD 306 transfers the data to the block buffer 406. This processing continues until a block count is reached. In an illustrative embodiment, a double buffer scheme is used in order to support high definition (HD) performance.
  • HD high definition
  • picture-level processing from the sequence level down to the macroblock level, including the sequence headers, picture headers, time stamps, and buffer management, are performed directly and sequentially by the core processor 302.
  • the VLD 306 assists the core processor when a bit-field in a header is to be decoded.
  • picture level processing does not overlap with slice level (macroblock) decoding.
  • some slice level or macroblock decoding processes may be performed concurrently with the picture level processing.
  • the macroblock level decoding is the main video decoding process. In an illustrative embodiment of the present invention it occurs within a direct execution loop. In such an embodiment, hardware blocks VLD 306, IQ 308, IDCT module 309, pixel filter 310, motion compensation module 312 (and possibly loop filter 313) are all involved in the decoding loop.
  • the core processor 302 controls the loop by polling the status of each of the hardware blocks involved. In an illustrative embodiment of the present invention, the actions of the various hardware blocks are arranged in an execution pipeline.
  • the pipeline scheme aims to achieve maximum utilization of the core processor 302.
  • FIG. 9 is a chart representing a decoding pipeline according to an illustrative embodiment of the present invention. The number of pipeline stages may vary depending on the target applications. Due to the selection of hardware elements that comprise the pipeline, the pipeline architecture of the present invention can accommodate substantially any existing or future compression algorithms that fall into the general class of DCT- based, variable-length coded, block-motion compensated algorithms.
  • FIG. 9 represent the decoding functions performed as part of the pipeline according to an exemplary embodiment.
  • Variable length decoding 900 is perfonned by VLD 306.
  • Run length/inverse scan/IQ/mismatch 902 are functions performed by IQ module 308.
  • IDCT operations 904 are performed by IDCT module 309.
  • Pixel filter reference fetch 906 and pixel filter reconstruction 908 are performed by pixel filter 310.
  • Motion compensation reconstruction 910 is perfonned by motion compensation module 312.
  • the columns of FIG. 9 represent the pipeline stages.
  • the designations MBj, MBi +1 , MBi +2 , etc. represent the i ⁇ macroblock in a data stream, the i+l st macroblock in the data stream, the i+2 nd macroblock, and so on.
  • variable length decoding 900 is performed on MBi.
  • Exploded view 920 of the variable length decoding function 900 demonstrates how functions are divided between the core processor 302 and the VLD 306 during this stage, according to one embodiment of the present invention. Exploded view 920 shows that during stage x 912, the core processor 302 decodes the macroblock header of MB;.
  • the VLD 306 assists the core processor 302 in the decoding of macroblock headers.
  • the core processor 302 also reconstructs the motion vectors of MBj, calculates the address of the pixel filter reference fetch for MB ⁇ , performs pipeline flow control and checks the status of IQ module 308, IDCT module 309, pixel filter 310 and motion compensator 312 during stage x 912.
  • the hardware blocks operate concunently with the core processor 302 while decoding a series of macroblocks.
  • the core processor 302 controls the pipeline, initiates the decoding of each macroblock, and controls the operation of each of the hardware accelerators.
  • the core processor firmware checks the status of each of the hardware blocks to determine completion of previously assigned tasks and checks the buffer availability before advancing the pipeline.
  • the VLD 306 also decodes the macroblock coefficients of MB; during stage x. Block coefficient VLC decoding is not started until the core processor 302 decodes the whole macroblock header. Note that the functions listed in exploded view 920 are performed during each stage of the pipeline of FIG. 9, even though, for simplicity's sake, they are only exploded out with respect to stage x 912.
  • stage x+1 914 the inverse quantizer 308 works on MBi (function 902) while variable length decoding 900 is perfonned on the next macroblock, MBj +1 .
  • stage x+1 914 the data that the inverse quantizer 308 works on are the quantized
  • the pixel filter reference data is fetched for MBi (function 906) using the pixel filter reference fetch address calculated by the core processor 302 during stage x 912. Then, at stage x+2 916, the IDCT module 309 performs IDCT operations 904 on the MBj DCT coefficients that were output by the inverse quantizer 308 during stage x+1.
  • the pixel filter 310 perfonns pixel filtering 908 for MBi using the pixel filter reference data fetched in stage x+1 914 and the motion vectors reconstructed by the core processor 302 in stage x 912. Additionally at stage x+2 916, the inverse quantizer 308 works on MBi + i (function 902), the pixel filter reference data is fetched for MBj +1 (function 906), and variable length decoding 900 is perfonned on MB; +2 .
  • the motion compensation module 312 performs motion compensation reconstruction 910 on MB; using decoded difference pixel information produced by the IDCT module 309 (function 904) and pixel prediction data produced by the pixel filter 310 (function 908) in stage x+2 916.
  • the IDCT module 309 performs IDCT operations 904 on MB i+1
  • the pixel filter 310 performs pixel filtering 908 for MBi + i
  • the inverse quantizer 308 works on MBi + (function 902)
  • the pixel filter reference data is fetched for MBj +2 (function 906)
  • variable length decoding 900 is performed on MBi +3 .
  • the pipeline of FIG. 9 shows just four pipeline stages, in an illustrative embodiment of the present invention, the pipeline includes as many stages as is needed to decode a complete incoming data stream with adequate performance.
  • the functions of two or more hardware modules are combined into one pipeline stage, and the macroblock data is processed by all the modules in that stage sequentially.
  • IDCT operations for a given macroblock are perfonned during the same pipeline stage as IQ operations.
  • the IDCT module 309 waits idle until the inverse quantizer 308 finishes, and the inverse quantizer 308 becomes idle when the IDCT operations start.
  • This embodiment will have a longer processing time for the "packed" pipeline stage, assuming the same performance of each individual function. Therefore, in an illustrative embodiment of the present invention, the packed pipeline stage is used only in non-demanding decoding tasks such SD (standard definition) or SIF (standard interchange format) size decoding applications.
  • the above-described macroblock-level pipeline advances stage-by-stage.
  • the pipeline advances after all the tasks in the current stage are completed.
  • the time elapsed in one macroblock pipeline stage will be referred to herein as the macroblock (MB) time.
  • the MB time is not a constant and varies from stage to stage. It depends on the encoded bitstream characteristics and possibly other factors, and is determined by the bottleneck module, which is the one that finishes last in that stage. Any module, including the core processor 302 itself, could be the bottleneck from stage to stage, and it is not pre- detennined at the beginning of each stage.
  • the bottleneck time is reduced by means of firmware control, improving the throughput and directly contributing to performance enhancement.
  • each module including the core processor 302 has a defined and predetermined task or group of tasks.
  • the maximum number of clock cycles needed for each module to decode decoding of a specific, e.g. worst case, stream can be predetermined.
  • the macroblock time for each module is substantially constant for streams with a given set of characteristics. Therefore, in an illustrative embodiment of the present invention, the hardware acceleration pipeline is optimized by hardware balancing each module in the pipeline according to the compression format of the data stream.
  • the main video decoding operations occur within a direct execution loop with polling of the accelerator functions.
  • the coprocessor/accelerators operate concunently with the core processor while decoding a series of macroblocks.
  • the core processor 302 controls the pipeline, initiates the decoding of each macroblock, and controls the operation of each of the accelerators.
  • firmware checks the status of each of the accelerators to determine completion of previously assigned tasks. In the event that the firmware gets to this point before an accelerator module has completed its required tasks, the firmware polls for completion. This is appropriate, since the pipeline cannot proceed efficiently until all of the pipeline elements have completed the cunent stage, and an interrupt driven scheme would be less efficient for this purpose.
  • Each hardware module 306, 308, 309, 310, 312, 313, 315 is independently controllable by the core processor 302.
  • the core processor 302 drives a hardware module by issuing a certain start command after checking the module's status.
  • the core processor 302 issues the start command by setting up a register in the hardware module.
  • FIG. 10 is a flowchart representing a macroblock decoding loop according to an illustrative embodiment of the present invention.
  • FIG. 10 depicts the decoding of one video picture.
  • the loop of slice/macroblock level decoding pipeline confrol is fully synchronous.
  • the core processor 302 retrieves a macroblock to be decoded from system memory 110.
  • the core processor starts all the hardware modules except the VLD 306.
  • the core processor 302 decodes the macroblock header with the help of the VLD 306.
  • the core processor 302 starts the VLD 306 for block coefficient decoding.
  • the core processor 302 calculates motion vectors and memory addresses, such as the pixel filter reference fetch address, controls buffer rotation and performs other housekeeping tasks.
  • decision box 1050 if the picture is decoded, the process is complete. If the picture is not decoded, the core processor 302 retrieves the next macroblock, and the process continues as shown by step 1000.
  • the core processor 302 interprets the bits decoded (with the help of the VLD 306) for the appropriate headers and sets up and coordinates the actions of the hardware modules.
  • the core processor 302 also controls and coordinates the actions of each hardware module.
  • the core processor configures the hardware modules to operate in accordance with the encoding/decoding format of the data stream being decoded by providing operating parameters to the hardware modules.
  • the parameters include but are not limited to (using MPEG-2 as an example) the cbp pattern used by the VLD 306 to decode the macroblock coefficients, the quantization scale used by the IQ module 308 to perfonn inverse quantization, motion vectors used by the pixel filter 309 and motion compensation module 310 to reconstruct the macroblocks, and the working buffer address(es) in decoder memory 316.
  • Each hardware module 306, 308, 309, 310, 312, 313, 315 performs the specific processing as instructed by the core processor 302 and sets up its status properly in a status register as the task is being executed and when it is done.
  • Each of the modules has or shares a status register that is polled by the core processor to detennine the module's status.
  • Each hardware module is assigned a set of macroblock buffers in decoder memory 316 for processing purposes.
  • Each hardware module signals the busy/available status of the working buffer(s) associated with it so that the core processor 302 can properly coordinate the processing pipeline.
  • the hardware accelerator modules 306, 308, 309, 319, 312, 313, 315 generally do not communicate with each other directly.
  • the accelerators work on assigned areas of decoder memory 316 and produce results that are written back to decoder memory 316, in some cases to the same area of decoder memory 316 as the input to the accelerator.
  • the accelerators in some cases may bypass the decoder memory 316 and pass data between themselves directly.
  • Software codes from other sources are ported to the decoding system 300 by analyzing the code to isolate those functions that are amenable to acceleration, such as variable-length decoding, run-length coding, inverse scanning, inverse quantization, transform, pixel filter, motion compensation, deblocking filter, and display format conversion, and replacing those functions with equivalent functions that use the hardware accelerators in the decoding system 300. All other video decoding software is compiled to run directly on the core processor.
  • FIG. 11 is a functional block diagram of a digital video decoding system 1100 according to an illustrative embodiment of the present invention.
  • Video decoding system 1100 is similar to the video decoding system 300 shown in FIG. 3, but includes two variable-length decoders, VLDo 1110 and VLDi 1120.
  • the other elements of FIG. 11 are equivalent to the elements shown and described with respect to FIG. 3.
  • both of the variable-length decoders 1110 and 1120 are programmable to decode bitsfreams of a plurality of compression/decompression standards.
  • each of the variable-length decoders 1110 and 1120 has a code RAM to hold VLC tables for various video coding fonnats.
  • variable-length decoder 1110 and 1120 is programmable to operate according to a plurality of compression/decompression standards, and the other variable-length decoder is hardwired to efficiently perform decoding according to one or more particular compression standards, such as MPEG-2 HD.
  • both of the variable-length decoders 1110 and 1120 are hardwired to efficiently perform decoding according to one or more particular compression standards.
  • one or both of the two VLDs 1110 and 1120 is hardwired to decode bitsfreams according to one or more particular standards and can also be optionally set as a programmable VLD engine, with a code RAM to hold VLC tables for other video coding formats.
  • the VLD includes a hard-coded coefficient decoder and a hard-coded code look-up table.
  • the two VLD engines 1110 and 1120 are controlled independently by the core processor 302, and either one or both of them will be employed at any given time, depending on the application.
  • the two variable-length decoders 1100 and 1110 are employed concunently to decode the same bitstream.
  • the two variable-length decoders are used in an alternating fashion to decode incoming macroblocks. That is VLDo 1110 is used to decode a first macroblock, VLDi 1120 is used to decode a second macroblock, VLDo 1100 is used to decode the third macroblock, and so on.
  • VLDo 1110 is used to decode a first macroblock
  • VLDi 1120 is used to decode a second macroblock
  • VLDo 1100 is used to decode the third macroblock, and so on.
  • two rows of a video frame are decoded concunently, with one row being decoded by one VLD, and the other row being decoded by the other VLD.
  • FIG. 12 is a chart representing a decoding pipeline according to an illustrative embodiment of the present invention.
  • the rows of FIG. 12 represent the functions performed by the core decoder processor
  • Row 1200 shows the functions performed by the core processor 302
  • row 1202 shows the functions perfonned by VLD 0 1110
  • row 1204 shows the functions performed by VLDi 1120.
  • the columns of FIG. 12 represent the pipeline stages.
  • the variable-length decoding of each macroblock is performed by one of the two VLDs 1110 and 1120 over two decoding stages.
  • the assigned VLD assists the core processor 302 in decoding the macroblock header.
  • the same VLD perfonns macroblock coefficient decoding for the same macroblock, while the other VLD assists the core processor 302 in decoding the macroblock header of a macroblock in a different row.
  • the decoder processor 302 performs macroblock header decoding on the first macroblock of rowo (rowo, columno).
  • VLDo 1110 assists the core processor 302 in the decoding of the macroblock header of the same macroblock (rowo, columno) data.
  • the decoder processor 302 performs macroblock header decoding on the first macroblock of rowi (rowi, columno), and VLDi 1120 assists the core processor 302 in the decoding of the macroblock header of the same macroblock
  • VLDo 1110 performs macroblock coefficient decoding for the first macroblock of rowo (rowo, columno), whose header was decoded by VLDo in stage x.
  • the decoder processor 302 performs macroblock header decoding on the second macroblock of rowo (rowo, columni), and VLDo 1110 assists the core processor 302 in the decoding of the macroblock header of the same macroblock (rowo, columni) data.
  • VLDi 1120 performs macroblock coefficient decoding for the first macroblock of row 1 (rowi, columno), whose header was decoded by VLDi in stage x+1. Decoding continues in this manner.
  • variable length decoder that is working on a given macroblock does not wait for the next stage after assisting the core processor 302 in decoding the macroblock header. Rather, when the decoding of the macroblock header is complete, the variable-length decoder begins decoding the macroblock coefficients for that macroblock, regardless of whether or not the next stage is ready to begin. This process will be refened to as cycle stealing.
  • FIG. 13 is a chart representing a dual-row decoding pipeline employing cycle stealing according to an illustrative embodiment of the present invention.
  • the rows of FIG. 13 represent the decoding functions performed as part of the pipeline, according to an exemplary embodiment of the present invention.
  • the functions include core processor operations 1300, variable-length decoding performed by VLDo 1302, variable-length decoding perfonned by VLDi 1304, inverse quantizer operations 1306, IDCT operations 1308, pixel filter reference fetch 1310, pixel filter reconstruction 1312, motion compensation 1314 and DMA operations 1316.
  • the columns of FIG. 13 represent the pipeline stages.
  • the designation (i, j) denotes the macroblock coordinates, i.e., the j & MB in the 1 th row.
  • the core processor 302 and VLDo 1110 work on MBo,o (MBo in rowo). Note that, first, the core processor 302 perfonns macroblock header decoding with the assistance of VLDo 1110. When the macroblock header is decoded, the core processor 302 continues perfonning other tasks, while VLDo 1110 begins decoding the block coefficients of MB 0> o- When the core processor 302 completes the tasks that it is performing with respect to MBo , o , the core processor 302 initiates stage 2, regardless of whether VLDo 1110 has finished decoding the block coefficients of MBo,o- In an alternative embodiment of the present invention, after assisting the core processor 302 with decoding the macroblock header, VLDo 1110 waits until stage 2 to begin decoding the block coefficients of MBo , o, as depicted in FIG. 12.
  • stage 2 the core processor 302 and VLDi 1120 work on MB ⁇ , o (MB 0 in rowi).
  • the core processor 302 performs macroblock header decoding on MBj . , 0 with the assistance of VLDi 1120.
  • the core processor 302 continues perfonning other tasks while VLDi 1120 begins decoding the block coefficients of MB ⁇ o.
  • VLDo 1110 did not finish decoding the block coefficients of MBo,o in stage 1, it (VLDo 1110) continues to do so in stage 2.
  • VLDo 1110 waits until stage 2 to begin decoding the block coefficients of MBo , o-
  • the core processor 302 polls VLDo 1110 to see if it is done decoding the block coefficients of
  • VLDo 1110 is done with MBo,o
  • the core processor 302 initiates stage 3, regardless of whether VLDi 1120 has finished decoding the block coefficients of
  • VLDo 1110 is not yet finished decoding the block coefficients of MB 0 ,o
  • the core processor 302 waits until VLDo 1110 is finished with MBo,o and initiates stage 3 at that time, again, regardless of whether VLDi 1120 has finished decoding the block coefficients of MB ⁇ ,o- '
  • stage 3 the core processor 302 and VLDo 1110 work on MBo, ⁇ (MBi in rowo) as described above with respect to stages 1 and 2. Also in stage 3, IQ module
  • the data that the inverse quantizer 308 works on are the quantized DCT coefficients of MBo , o , extracted from the data stream by VLDo 1110 during stage 2. Additionally in stage 3, VLDi 1120 continues decoding the block coefficients of MB ⁇ ; o, if the decoding was not completed in stage 2.
  • the core processor 302 polls VLDi to see if it is done decoding the block coefficients of MB ⁇ ( o.
  • the core processor 302 also polls IQ module 308 to see if it is done operating on MBo, ⁇ . If VLDi 1120 is done with MBo,o, and IQ module 308 is done with MB 0> ⁇ , the core processor 302 initiates stage 4, regardless of whether VLDo 1110 has finished decoding the block coefficients of MBo , ⁇ . If either VLDi 1120 or IQ module 308 is not yet finished, the core processor 302 waits until VLDi 1120 and IQ module 308 are both finished and initiates stage 4 at that time.
  • the pixel filter reference data is fetched for MBo,o (function 910), using the pixel filter reference fetch address calculated by the core processor 302 during stage 1.
  • the core processor 302 also polls the pixel filter 310 for completion prior to initiating stage 4.
  • stage 4 the core processor 302 works on MB 1; ⁇ (MBi in rowi), variable- length decoding is initiated on MBi by VLDi 1120, IQ module 308 operates on MB ⁇ o, and the pixel filter reference data is fetched for MB 1; o (function 910). Also in stage 4, IDCT module 309 performs the inverse transform on the MBo , o coefficients produced by the IQ module 308 in stage 3, and the pixel filter 310 performs pixel filtering 912 for MB 0, o, using the pixel filter reference data fetched in stage 3 and the motion vectors reconstructed by the core processor 302 in stage 1.
  • VLDo 1110 continues decoding the block coefficients of MBo , ⁇ if the decoding was not completed in stage 3.
  • the core processor 302 completes its tasks with respect to the core processor 302 polls VLDo 1110, IQ module 308, IDCT module 309 and pixel filter 310 to see if they have completed their present tasks. If the polled modules have completed their tasks, the core processor 302 initiates stage 5. If any of the polled modules is not yet finished, the core processor waits until they are all finished and initiates stage 5 at that time.
  • stage 5 the core processor 302 works on MBo, 2 (MB 2 in rowo), variable- length decoding is initiated on MB 0)2 by VLD 0 1110, IQ module 308 operates on MBo.i, IDCT module 309 operates on the MB 1; o coefficients, the pixel filter reference data is fetched for MBo , ⁇ (function 910), and the pixel filter 310 perfonns pixel filtering 912 for MB ⁇ ; o- Also in stage 5, the motion compensation module 312 performs motion compensation reconstruction 914 on MBo,o, using decoded difference pixel information produced by the IDCT module 309 (function 908) and pixel prediction data produced by the pixel filter 310 (function 912) in stage 4 616.
  • VLDi 1120 continues decoding the block coefficients of MBi if the decoding was not completed in stage 4.
  • the core processor 302 polls VLDi 1120, IQ module 308, IDCT module 309, pixel filter 310 and motion compensation module 312 to see if they have completed their present tasks. If the polled modules have completed their tasks, the core processor 302 initiates stage 6. If any of the polled modules is not yet finished, the core processor waits until they are all finished and initiates stage 6 at that time.
  • the core processor 302 works on MB ⁇ ;2 (MB 2 in rowi), variable- length decoding is initiated on MB ⁇ >2 by VLDi 1120, IQ module 308 operates on MB 1; ⁇ , IDCT module 309 operates on the MBo.i coefficients, the pixel filter reference data is fetched for MBi (function 910), the pixel filter 310 perfonns pixel filtering
  • the DMA engine 304 places the result of the motion compensation performed with respect to MBo,o in system memory 110. Additionally in stage 5, VLDo 1110 continues decoding the block coefficients of MB 0,2 if the decoding was not completed in stage 5.
  • the core processor 302 polls VLDi 1120, IQ module 308, IDCT module 309, pixel filter 310, motion compensation module 312 and DMA engine 304 to see if they have completed their present tasks. If the polled modules have completed their tasks, the core processor 302 initiates stage 7. If any of the polled modules is not yet finished, the core processor waits until they are all finished and initiates stage 7 at that time.
  • the decoding pipeline described above with respect to FIG. 13 continues as long as there are further macroblocks in the data stream to decode.
  • the dual-row decoding pipeline demonstrated in FIG. 13 can be implemented in any type of decoding scheme (including, e.g., audio decoding) employing any combination of acceleration modules.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Discrete Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Compression Or Coding Systems Of Tv Signals (AREA)
  • Compression, Expansion, Code Conversion, And Decoders (AREA)

Abstract

La présente invention concerne un système de décodage vidéo comprenant un accélérateur de décodeur à longueur programmable. Le système comprend un processeur de décodeur et un accélérateur de décodage à longueur variable. L'accélérateur de décodage à longueur variable est couplé au processeur de décodeur et effectue des opérations de décodage à longueur variable sur le code à longueur variable présent dans le flux de données vidéo. L'accélérateur de décodage à longueur variable est capable de décoder le code à longueur variable conformément à n'importe quelle méthode de décodage. Dans une forme de réalisation, le décodeur à longueur variable comprend une pluralité de tableaux de code qui sont stockés en mémoire et un registre de sélection de tableaux de code qui est programmable pour imposer le tableau de code qui doit être utilisé pour le décodage du code à longueur variable. Dans une forme de réalisation, le système de décodage comprend deux accélérateurs de décodage à longueur variable.
PCT/US2003/009896 2002-04-01 2003-04-01 Systeme de decodage video comprenant un decodeur a longueur variable programmable WO2003085494A2 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US36914402P 2002-04-01 2002-04-01
US60/369,144 2002-04-01

Publications (2)

Publication Number Publication Date
WO2003085494A2 true WO2003085494A2 (fr) 2003-10-16
WO2003085494A3 WO2003085494A3 (fr) 2003-12-18

Family

ID=28791929

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2003/009896 WO2003085494A2 (fr) 2002-04-01 2003-04-01 Systeme de decodage video comprenant un decodeur a longueur variable programmable

Country Status (1)

Country Link
WO (1) WO2003085494A2 (fr)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1562383A3 (fr) * 2004-02-09 2005-08-17 Broadcom Corporation Décodeur vidéo multistandard
WO2006105544A1 (fr) * 2005-03-30 2006-10-05 Intel Corporation Decodeur a longueur variable multinorme a accelerateur materiel
JP4891335B2 (ja) * 2005-12-09 2012-03-07 エヌヴィディア コーポレイション ハードウェア多標準対応ビデオデコーダ装置

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100335138B1 (ko) * 1998-12-30 2002-11-27 엘지정보통신주식회사 비디오코더의가변부호화기및이를이용한코딩방법

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1562383A3 (fr) * 2004-02-09 2005-08-17 Broadcom Corporation Décodeur vidéo multistandard
US7720294B2 (en) 2004-02-09 2010-05-18 Broadcom Corporation Unified decoder architecture
WO2006105544A1 (fr) * 2005-03-30 2006-10-05 Intel Corporation Decodeur a longueur variable multinorme a accelerateur materiel
JP2008537392A (ja) * 2005-03-30 2008-09-11 インテル コーポレイション ハードウェア・アクセラレータを備えたマルチ標準可変復号器
KR100985361B1 (ko) * 2005-03-30 2010-10-04 인텔 코포레이션 데이터 처리를 위한 장치, 시스템 및 방법
US8705632B2 (en) 2005-03-30 2014-04-22 Intel Corporation Decoder architecture systems, apparatus and methods
JP4891335B2 (ja) * 2005-12-09 2012-03-07 エヌヴィディア コーポレイション ハードウェア多標準対応ビデオデコーダ装置

Also Published As

Publication number Publication date
WO2003085494A3 (fr) 2003-12-18

Similar Documents

Publication Publication Date Title
US8913667B2 (en) Video decoding system having a programmable variable-length decoder
USRE48845E1 (en) Video decoding system supporting multiple standards
US8005147B2 (en) Method of operating a video decoding system
US6963613B2 (en) Method of communicating between modules in a decoding system
US7403564B2 (en) System and method for multiple channel video transcoding
US8213511B2 (en) Video encoder software architecture for VLIW cores incorporating inter prediction and intra prediction
US9131240B2 (en) Video decoding method and apparatus which uses double buffering
US6771196B2 (en) Programmable variable-length decoder
EP1440585A2 (fr) Decodeur a longueur variable ameliore
US7095341B2 (en) Programmable variable-length decoder
KR20030005199A (ko) 스케일링 가능한 계산 복잡성 비디오 및 정지 이미지복호를 위한 근사적 역 이산 코사인 변환
EP1351512A2 (fr) Système de décodage vidéo adapté à plusieurs standards
KR20090020460A (ko) 비디오 디코딩 방법 및 장치
WO2003085494A2 (fr) Systeme de decodage video comprenant un decodeur a longueur variable programmable
EP1351513A2 (fr) Procédé de fonctionnement d'un système de décodage vidéo
US10778980B2 (en) Entropy decoding apparatus with context pre-fetch and miss handling and associated entropy decoding method
EP1351511A2 (fr) Procédé de communication entre des modules dans un système de décodage vidéo
US20070192393A1 (en) Method and system for hardware and software shareable DCT/IDCT control interface
Bhanja et al. Hardware implementation of data compression
Li et al. A decoder architecture for advanced video coding standard
Lakshmish et al. Efficient Implementation of VC-1 Decoder on Texas Instrument's OMAP2420-IVA
Li Optimization of queueing performance and design variables in a single-bus shared-memory system with application to MPEG-2 video decoder system

Legal Events

Date Code Title Description
AL Designated countries for regional patents

Kind code of ref document: A2

Designated state(s): AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LU MC NL PT RO SE SI SK TR

121 Ep: the epo has been informed by wipo that ep was designated in this application
DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
122 Ep: pct application non-entry in european phase
点击 这是indexloc提供的php浏览器服务，不要输入任何密码和下载