这是indexloc提供的服务,不要输入任何密码
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd"> <html> <head> <meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1"> <meta name="keywords" content="SBMI2015, Scene Background Initialization"> <title>Scene Background Initialization (SBI) dataset</title> </head> <body> <span style='font-family:"Calibri";color:navy;'> <b><h2 align="center">Scene Background Initialization (SBI) dataset</h2></b></p> The <a target="_blank" href=http://sbmi2015.na.icar.cnr.it/SBIdataset.html>SBI dataset</a> has been assembled in order to move the first steps towards evaluating and comparing the results of background initialization algorithms and adopted for the <a target="_blank" href=http://sbmi2015.na.icar.cnr.it>SBMI2015 Workshop</a> and the <a target="_blank" href=https://www.sciencedirect.com/journal/pattern-recognition-letters/vol/96> Special Issue of Pattern Recognition Letters Journal on Scene Background Modeling and Initialization</a> (2016). A description and some results can be found in <a href="#[1]">our work [1]</a>. The dataset includes: <p> A) a <a href="#[Dataset]">dataset of 14 image sequences</a> and the corresponding 14 <a href="#[Dataset]">ground truth backgrounds</a> <b>(updated on January 25, 2016)</b>; <p> B) <a href="#[Metrics]">Matlab scripts</a> <b>(updated on July 26, 2016)</b> for evaluating background initialization results.</p> <p><a href="#[Results]">Results of methods reported in</a> <a href="#[2]">[2]</a> are also available <b>(updated on November 30, 2016).</b></p> <!-- Any other researcher involved in background initialization is invited to conduct and report results of quantitative evaluation of their methods using the above material. <b>Please, check for signaled updates,</b> as we are extending the provided material with the aim of enhancing the proposed quantitative analysis. To this end, any suggestion is strongly appreciated. </p>--> <p><a name=[Dataset]><div align="left"><h3>A) Dataset of 14 image sequences and corresponding ground truth backgrounds</h3></div></a></p> The <b>14 image sequences</b> have been extracted by original publicly available sequences that are frequently used in the literature to evaluate background initialization algorithms:</p> <div align="center"> <span align="center"> <table align="center" border="1"> <tr><b> <td><b>Name</b></td> <td><b>Dataset</b></td> <td><b>Original<br>frames</b></td> <td><b>Used<br>frames</b></td> <td><b>Original<br>resolution</b></td> <td><b>Final<br>resolution</b></td> <td><b>Brief description<br>and issues</b></td> </b></tr> <tr><td>Board</td><td><a target="_blank" href=http://www.diegm.uniud.it/fusiello/demo/bkg/>PBI</a></td><td>0-227 </td><td>0-227 </td><td>200x164 </td><td>200x164</td><td>Man moving in front of a dashboard, with mild shadows</td></tr> <tr><td>Candela_m1.10</td><td><a target="_blank" href=http://www.multitel.be/image/research-development/research-projects/candela/abandon-scenario.php>Candela</a></td><td> 0-855</td><td> 85-435</td><td> 352x288</td><td> 352x288</td><td>Man entering and leaving a room, abandoning a bag for most of the frames</td></tr> <tr><td>CAVIAR1</td><td><a target="_blank" href=http://groups.inf.ed.ac.uk/vision/CAVIAR/CAVIARDATA1/>CAVIAR</a></td><td> 0-725</td><td> 115-724</td><td> 384x288</td><td> 384x256</td><td>People slowly walking along a corridor, with mild shadows</td></tr> <tr><td>CAVIAR2</td><td><a target="_blank" href=http://groups.inf.ed.ac.uk/vision/CAVIAR/CAVIARDATA1/>CAVIAR</a></td><td> 0-1500</td><td> 900-1360</td><td> 384x288</td><td> 384x256</td><td>People entering and leaving a store, standing only for few frames</td></tr> <tr><td>CaVignal</td><td><a target="_blank" href=http://www.diegm.uniud.it/fusiello/demo/bkg/>PBI</a></td><td>0-257 </td><td>0-257 </td><td>200x136 </td><td>200x136</td><td>Man standing for most of the frames and then moving</td></tr> <tr><td>Foliage</td><td><a target="_blank" href=http://www.diegm.uniud.it/fusiello/demo/bkg/>PBI</a></td><td>0-399 </td><td>6-399 </td><td>200x148 </td><td>200x144</td><td>Parked cars occluded by big waving leaves</td></tr> <tr><td>Hall&Monitor</td><td><a target="_blank" href=http://www.ics.forth.gr/cvrl/demos/NEMESIS/hall_monitor.mpg>COST 211</a></td><td>0-299</td><td> 4-299</td><td> 352x240</td><td> 352x240</td><td>Walking person and abandoned bag in the same image region for most of the frames</td></tr> <tr><td>HighwayI</td><td><a target="_blank" href=http://cvrr.ucsd.edu/aton/shadow/index.html>ATON</a></td><td> 0-439</td><td> 0-439</td><td> 320x240</td><td> 320x240</td><td>Fast motion of cars along a highway, with strong shadows and small camera jitter</td></tr> <tr><td>HighwayII</td><td><a target="_blank" href=http://cvrr.ucsd.edu/aton/shadow/index.html>ATON</a></td><td> 0-499</td><td> 0-499</td><td> 320x240</td><td> 320x240</td><td>Fast motion of cars along a highway, with strong shadows and small camera jitter</td></tr> <tr><td>HumanBody2</td><td><a target="_blank" href=https://www.ecse.rpi.edu/~cvrl/humanbody/>RPI ISL</a></td><td> 0-898</td><td> 70-810</td><td> 320x240</td><td> 320x240</td><td>People quickly walking indoor, with mild shadows</td></tr> <tr><td>IBMtest2</td><td><a target="_blank" href=http://www.research.ibm.com/peoplevision/performanceevaluation.html>IBM</a></td><td>0-1750 </td><td>1027-1117 </td><td>320x240 </td><td>320x240</td><td>People quickly walking along indoor corridors</td></tr> <tr><td>People&Foliage</td><td><a target="_blank" href=http://www.diegm.uniud.it/fusiello/demo/bkg/>PBI</a></td><td>0-349 </td><td>0-340 </td><td>320x240 </td><td>320x240</td><td>Parked cars occluded by moving people and big waving leaves</td></tr> <tr><td>Snellen</td><td><a target="_blank" href=http://www.diegm.uniud.it/fusiello/demo/bkg/>PBI</a></td><td> 0-333</td><td> 0-320</td><td> 146x150</td><td> 144x144</td><td>Stationary Snellen chart occluded and shadowed by big waving leaves</td></tr> <tr><td>Toscana</td><td><a target="_blank" href=http://www.mpi-inf.mpg.de/~granados/>MPI Informatik</a></td><td> 0-5</td><td> 0-5</td><td> 2272x1704</td><td> 800x600</td><td>Very few outdoor pictures of pedestrians taken at different times</td></tr> </table> </span> </div> <p>The subsets of used frames have been selected in order to avoid the inclusion into the testing sequences of empty frames (frames not including foreground objects), while the final resolution has been chosen in order to avoid problems in the computation of boundary patches for block-based methods.</p> The <b>ground truths</b> have been manually obtained by choosing one of the sequence frames free of foreground objects (not included into the subsets of used frames), by stitching together empty background regions from different sequence frames, or by temporal median of background regions. <span align="center"> <table> <tr> <td>1.</td> <td><i>Board</i><br> (<a target="_blank" href=http://www.diegm.uniud.it/fusiello/demo/bkg/>PBI data set</a>)<br></td> <td align="center"><img src="./MODLab/BckgInit/Sequences/Board_000100.png" border="0" height="80"><br> Sequence: <a href=./MODLab/BckgInit/Sequences/Board.zip>PNG video frames (zipped)</a><br></td> <td align="center"><img src="./MODLab/BckgInit/GT/GT_Board.png" border="0" height="80"><br> Ground truth: <a href=./MODLab/BckgInit/GT/GT_Board.png>Background</a></td> <tr> <td>2.</td> <td><i>Candela_m1.10</i><br> (<a target="_blank" href=http://www.multitel.be/image/research-development/research-projects/candela/abandon-scenario.php>Candela data set</a>)<br></td> <td align="center"><img src="./MODLab/BckgInit/Sequences/Candela_m1.10_000100.png" border="0" height="80"><br> Sequence: <a href=./MODLab/BckgInit/Sequences/Candela_m1.10.zip>PNG video frames (zipped)</a><br></td> <td align="center"><img src="./MODLab/BckgInit/GT/GT_Candela_m1.10.png" border="0" height="80"><br> Ground truth: <a href=./MODLab/BckgInit/GT/GT_Candela_m1.10.png>Background</a><br></td> <tr> <td>3.</td> <td><i>CAVIAR1</i><br> (<a target="_blank" href=http://groups.inf.ed.ac.uk/vision/CAVIAR/CAVIARDATA1/>CAVIAR Test Case Scenarios</a>)<br></td> <td align="center"><img src="./MODLab/BckgInit/Sequences/CAVIAR1_000100.png" border="0" height="80"><br> Sequence: <a href=./MODLab/BckgInit/Sequences/CAVIAR1.zip>PNG video frames (zipped)</a><br></td> <td align="center"><img src="./MODLab/BckgInit/GT/GT_CAVIAR1.png" border="0" height="80"><br> Ground truth: <a href=./MODLab/BckgInit/GT/GT_CAVIAR1.png>Background</a><br></td> <tr> <td>4.</td> <td><i>CAVIAR2</i><br> (<a target="_blank" href=http://groups.inf.ed.ac.uk/vision/CAVIAR/CAVIARDATA1/>CAVIAR Test Case Scenarios</a>)<br></td> <td align="center"><img src="./MODLab/BckgInit/Sequences/CAVIAR2_000100.png" border="0" height="80"><br> Sequence: <a href=./MODLab/BckgInit/Sequences/CAVIAR2.zip>PNG video frames (zipped)</a><br></td> <td align="center"><img src="./MODLab/BckgInit/GT/GT_CAVIAR2.png" border="0" height="80"><br> Ground truth: <a href=./MODLab/BckgInit/GT/GT_CAVIAR2.png>Background</a><br></td> <tr> <td>5.</td> <td><i>CaVignal</i><br> (<a target="_blank" href=http://www.diegm.uniud.it/fusiello/demo/bkg/>PBI data set</a>)<br></td> <td align="center"><img src="./MODLab/BckgInit/Sequences/CaVignal_000100.png" border="0" height="80"><br> Sequence: <a href=./MODLab/BckgInit/Sequences/CaVignal.zip>PNG video frames (zipped)</a><br></td> <td align="center"><img src="./MODLab/BckgInit/GT/GT_CaVignal.png" border="0" height="80"><br> Ground truth: <a href=./MODLab/BckgInit/GT/GT_CaVignal.png>Background</a></td> <tr> <td>6.</td> <td><i>Foliage</i><br> (<a target="_blank" href=http://www.diegm.uniud.it/fusiello/demo/bkg/>PBI data set</a>)<br></td> <td align="center"><img src="./MODLab/BckgInit/Sequences/Foliage_000100.png" border="0" height="80"><br> Sequence: <a href=./MODLab/BckgInit/Sequences/Foliage.zip>PNG video frames (zipped)</a><br></td> <td align="center"><img src="./MODLab/BckgInit/GT/GT_Foliage.png" border="0" height="80"><br> Ground truth: <a href=./MODLab/BckgInit/GT/GT_Foliage.png>Background</a></td> <tr> <td>7.</td> <td><i>Hall&Monitor</i><br> (<a target="_blank" href=http://www.ics.forth.gr/cvrl/demos/NEMESIS/hall_monitor.mpg>COST 211 data set</a>)<br></td> <td align="center"><img src="./MODLab/BckgInit/Sequences/HallAndMonitor_000100.png" border="0" height="80"><br> Sequence: <a href=./MODLab/BckgInit/Sequences/HallAndMonitor.zip>PNG video frames (zipped)</a>&nbsp;&nbsp;&nbsp;&nbsp;<br></td> <td align="center"><img src="./MODLab/BckgInit/GT/GT_HallAndMonitor.png" border="0" height="80"><br>Ground truth: <a href=./MODLab/BckgInit/GT/GT_HallAndMonitor.png>Background</a><br></td> </tr> <tr> <td>8.</td> <td><i>HighwayI</i><br> (<a target="_blank" href=http://cvrr.ucsd.edu/aton/shadow/index.html>ATON data set</a>)<br></td> <td align="center"><img src="./MODLab/BckgInit/Sequences/HighwayI_000100.png" border="0" height="80"><br> Sequence: <a href=./MODLab/BckgInit/Sequences/HighwayI.zip>PNG video frames (zipped)</a><br></td> <td align="center"><img src="./MODLab/BckgInit/GT/GT_HighwayI.png" border="0" height="80"><br> Ground truth: <a href=./MODLab/BckgInit/GT/GT_HighwayI.png>Background</a></td> </tr> <tr> <td>9.</td> <td><i>HighwayII</i><br> (<a target="_blank" href=http://cvrr.ucsd.edu/aton/shadow/index.html>ATON data set</a>)<br></td> <td align="center"><img src="./MODLab/BckgInit/Sequences/HighwayII_000100.png" border="0" height="80"><br> Sequence: <a href=./MODLab/BckgInit/Sequences/HighwayII.zip>PNG video frames (zipped)</a><br></td> <td align="center"><img src="./MODLab/BckgInit/GT/GT_HighwayII.png" border="0" height="80"><br> Ground truth: <a href=./MODLab/BckgInit/GT/GT_HighwayII.png>Background</a></td> <tr> <td>10.</td> <td><i>HumanBody2</i><br> (<a target="_blank" href=https://www.ecse.rpi.edu/~cvrl/humanbody/>RPI ISL data set</a>)<br></td> <td align="center"><img src="./MODLab/BckgInit/Sequences/HumanBody2_000100.png" border="0" height="80"><br> Sequence: <a href=./MODLab/BckgInit/Sequences/HumanBody2.zip>PNG video frames (zipped)</a><br></td> <td align="center"><img src="./MODLab/BckgInit/GT/GT_HumanBody2.png" border="0" height="80"><br> Ground truth: <a href=./MODLab/BckgInit/GT/GT_HumanBody2.png>Background</a><br></td> <tr> <td>11.</td> <td><i>IBMtest2</i><br> (<a target="_blank" href=http://www.research.ibm.com/peoplevision/performanceevaluation.html>IBM Research - PeopleVision data set</a>)<br></td> <td align="center"><img src="./MODLab/BckgInit/Sequences/IBMtest2_000050.png" border="0" height="80"><br> Sequence: <a href=./MODLab/BckgInit/Sequences/IBMtest2.zip>PNG video frames (zipped)</a><br></td> <td align="center"><img src="./MODLab/BckgInit/GT/GT_IBMtest2.png" border="0" height="80"><br> Ground truth: <a href=./MODLab/BckgInit/GT/GT_IBMtest2.png>Background</a></td> <tr> <td>12.</td> <td><i>People&Foliage</i><br> (<a target="_blank" href=http://www.diegm.uniud.it/fusiello/demo/bkg/>PBI data set</a>)<br></td> <td align="center"><img src="./MODLab/BckgInit/Sequences/PeopleAndFoliage_000100.png" border="0" height="80"><br> Sequence: <a href=./MODLab/BckgInit/Sequences/PeopleAndFoliage.zip>PNG video frames (zipped)</a><br></td> <td align="center"><img src="./MODLab/BckgInit/GT/GT_PeopleAndFoliage.png" border="0" height="80"><br> Ground truth: <a href=./MODLab/BckgInit/GT/GT_PeopleAndFoliage.png>Background</a></td> <tr> <td>13.</td> <td><i>Snellen</i><br> (<a target="_blank" href=http://www.diegm.uniud.it/fusiello/demo/bkg/>PBI data set</a>)<br></td> <td align="center"><img src="./MODLab/BckgInit/Sequences/Snellen_000100.png" border="0" height="80"><br> Sequence: <a href=./MODLab/BckgInit/Sequences/Snellen.zip>PNG video frames (zipped)</a><br></td> <td align="center"><img src="./MODLab/BckgInit/GT/GT_Snellen.png" border="0" height="80"><br> Ground truth: <a href=./MODLab/BckgInit/GT/GT_Snellen.png>Background</a><br></td> <tr> <td>14.</td> <td><i>Toscana</i><br> (<a target="_blank" href=http://www.mpi-inf.mpg.de/~granados/>MPI Informatik data set</a>)<br></td> <td align="center"><img src="./MODLab/BckgInit/Sequences/Toscana_000003.png" border="0" height="80"><br> Sequence: <a href=./MODLab/BckgInit/Sequences/Toscana.zip>PNG video frames (zipped)</a><br></td> <td align="center"><img src="./MODLab/BckgInit/GT/GT_Toscana.png" border="0" height="80"><br> Ground truth: <a href=./MODLab/BckgInit/GT/GT_Toscana.png>Background</a><br></td> </table> </span> Click <a href=./MODLab/BckgInit/GT/GT.zip>here</a> to download all the 14 ground truths (zipped). <p><a name=[Metrics]><div align="left"><h3>B) Matlab scripts for evaluating background initialization results</h3></div></a></p> <a href=./MODLab/BckgInit/MATLAB/EvaluateBckgInit3.zip>Matlab scripts</a> <b>(updated on July 26, 2016)</b> are provided for evaluating results in terms of six metrics that include those used in the literature for background estimation. Denoting with GT an image containing the true background and with CB the estimated background image computed with a background initialization method, these six metrics are defined as follows: </p> <ol> <li><b>Average Gray-level Error (AGE)</b>: It is the average of the gray-level absolute difference between GT and CB images. Its values range in [0, L-1], where L is the maximum number of grey levels; the lower the AGE value, the better is the background estimate.</p> <li><b>Percentage of Error Pixels (pEPs)</b>: An error pixel (EP) is a pixel of CB whose value differs from the value of the corresponding pixel in GT by more than some threshold th (in the experiments the value th=20 has been suggested). pEPs is the ratio between the EPs and the number N of image pixels. Its values range in [0, 1]; the lower the pEPs value, the better is the background estimate. </p> <li><b>Percentage of Clustered Error Pixels (pCEPs)</b>: A clustered error pixel (CEP) is defined as any error pixel whose 4-connected neighbors are also error pixels. pCEPs is the ratio between the CEPs and the number N of image pixels. Its values range in [0,1]; the lower the pCEPs value, the better is the background estimate. </p> <li><b>Peak-Signal-to-Noise-Ratio (PSNR)</b>: It is defined as PSNR = 10 log<sub>10</sub>((L-1)<sup>2</sup>/MSE), where L is the maximum number of grey levels and MSE is the Mean Squared Error between GT and CB images. It assumes values in decibels; the higher the PSNR value, the better is the background estimate.</p> <li><b>Multi-Scale Structural Similarity Index (MS-SSIM)</b>: This is the metric defined by Z. Wang, E. P. Simoncelli and A. C. Bovik (<a target="_blank" href="https://www.google.it/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0ahUKEwiTqs3GnYLPAhVHvRoKHVffBacQFgghMAA&url=https%3A%2F%2Fece.uwaterloo.ca%2F~z70wang%2Fpublications%2Fmsssim.pdf&usg=AFQjCNFjzs2hQdG3ceghHA8mmwHpeaU1cQ">link</a>), that uses structural distortion as an estimate of the perceived visual distortion. It assumes values in [-1; 1]; the higher the value of MS-SSIM, the better is the estimated background.</p> <li><b>Color image Quality Measure (CQM)</b>: This is the metric recently proposed by Y. Yalman and I. Erturk (<a target="_blank" href="http://journals.tubitak.gov.tr/elektrik/issues/elk-13-21-2/elk-21-2-20-1111-11.pdf">link</a>), based on a reversible transformation of the YUV color space and on the PSNR computed in the single YUV bands. As for the PSNR, it assumes values in decibels; the higher the CQM value, the better is the background estimate.</p> </ol> </p>If you have problems downloading, please contact lucia.maddalena@cnr.it; if you use the SBI dataset, please cite <a href="#[1]">our works [1], [2]</a>. <p><h3><a name=[Results]><div align="left">Results of methods reported in</a> <a href="#[2]">[2]</a></h3></div></p> Here, we report all the results of the background initialization methods compared in <a href="#[2]">[2]</a> for each sequence of the SBI dataset, so that each new method can be easily compared with those considered in <a href="#[2]">[2]</a>. Table 1 reports average accuracy results obtained by the compared methods according to the adopted metrics, while Tables 2 through 15 report accuracy results of all the compared methods on each SBI sequence, as well as their average per sequence. In all the Tables, the best and the second best results for each metric and each sequence appear in <span style='Color:red'><b>red</b></span> and <b>blue</b>, respectively. </p> <div align="center"> <span align="center"> <table align="center" border="1"> <caption>Table 1: Average accuracy results of the compared methods on all sequences of the SBI dataset.</caption> <tr><td><b>Method</b></td> <td><b>AGE</b></td> <td><b>pEPs</b></td> <td><b>pCEPs</b></td> <td><b>MS-SSIM</b></td> <td><b>PSNR</b></td> <td><b>CQM</b></td></tr> <td>Mean </td><td>14.1944 </td><td>22.5150 </td><td>18.4428 </td><td>0.8737 </td><td>25.6980 </td><td>43.5839 </tr> <td>Color Median </td><td>10.3744 </td><td>13.4008 </td><td>10.5571 </td><td>0.8533 </td><td>28.0044 </td><td>42.4746 </tr> <td>MOG2 </td><td>14.3579 </td><td>4.0847 </td><td>2.8080 </td><td>0.8935 </td><td>25.9576 </td><td>38.1916</tr> <td>KNN </td><td>20.6968 </td><td>7.5118 </td><td>4.5180 </td><td>0.7595 </td><td>18.4701 </td><td>26.3836 </tr> <td>BE-AAPSA </td><td>11.4846 </td><td>12.5518 </td><td>10.0605 </td><td>0.9247 </td><td>27.8024 </td><td>41.8124</tr> <td>WS2006 </td><td>5.2885 </td><td>3.5335 </td><td><b>1.2118</b></td> </td><td>0.9349 </td><td>28.8791 </td><td>39.6334</tr> <td>IMBS-MT </td><td>4.2092 </td><td>3.8819 </td><td>2.2602 </td><td>0.9598 </td><td>33.4090 </td><td>44.9362 </tr> <td>LaBGen </td><td><span style='Color:red'><b>2.9945</b></td></td><td><span style='Color:red'><b>1.3972</b></td></td><td><span style='Color:red'><b>0.9246</b></td> </td><td><b>0.9764</b></td> </td><td><b>35.2028</b></td> </td><td><b>47.2947</b></td></tr> <td>RSL2011 </td><td>5.8228 </td><td>5.3511 </td><td>4.0186 </td><td>0.9172 </td><td>29.9272 </td><td>40.5713 </tr> <td>Photomontage </td><td>5.8238 </td><td>4.6952 </td><td>3.7274 </td><td>0.9334 </td><td>31.8573 </td><td>43.9038 </tr> <td>LRGeomCG </td><td>8.7644 </td><td>14.1305 </td><td>11.0810 </td><td>0.9302 </td><td>28.9596 </td><td>45.5625 </tr> <td>TMac </td><td>8.8685 </td><td>14.3577 </td><td>11.2884 </td><td>0.9284 </td><td>28.7507 </td><td>45.4125</tr> <td>SC-SOBS_1 </td><td><b>3.5023</b></td> </td><td>4.1508 </td><td>2.2295 </td><td><span style='Color:red'><b>0.9765</b></td> </td><td><span style='Color:red'><b>35.2723</b></td> </td><td><span style='Color:red'><b>50.1138</b></td></tr> <td>SC-SOBS_2 </td><td>4.6049 </td><td>4.7435 </td><td>2.5370 </td><td>0.9645 </td><td>32.2024 </td><td>45.7614 </tr> <td>BEWIS </td><td>3.8665 </td><td><b>2.4286</b></td> </td><td>1.4238 </td><td>0.9675 </td><td>32.0143 </td><td>44.3728 </tr> </table> </span> </div> <br> <div align="center"> <span align="center"> <table align="center" border="1"> <caption>Table 2: Accuracy results of all the compared methods on sequence <i>Board</i> and their Average. </caption> <tr><td><b>Method</b></td> <td><b>AGE</b></td> <td><b>pEPs</b></td> <td><b>pCEPs</b></td> <td><b>MS-SSIM</b></td> <td><b>PSNR</b></td> <td><b>CQM</b></td></tr> <td>Mean </td><td>24.9527 </td><td>45.4055 </td><td>37.7530 </td><td>0.6100 </td><td>17.9834 </td><td>44.0696 </tr> <td>Color Median </td><td>17.9861 </td><td>23.4360 </td><td>20.1433 </td><td>0.4549 </td><td>17.7707 </td><td>43.5605 </tr> <td>MOG2 </td><td>21.5981 </td><td>23.3689 </td><td>15.2652 </td><td>0.8433 </td><td>17.0541 </td><td>29.2805 </tr> <td>KNN </td><td>31.1259 </td><td>26.8963 </td><td>17.2561 </td><td>0.7734 </td><td>13.4368 </td><td>21.3434 </tr> <td>BE-AAPSA </td><td>21.6749 </td><td><span style='Color:red'><b>0.2904</b></td> </td><td>0.2459 </td><td>0.7705 </td><td>16.9839 </td><td>32.3073 </tr> <td>WS2006 </td><td>8.1210 </td><td>6.0945 </td><td>0.6799 </td><td>0.8007 </td><td>23.5534 </td><td>34.6697</tr> <td>IMBS-MT </td><td><span style='Color:red'><b>2.2537</b></td> </td><td><b>0.3201</b></td> </td><td><span style='Color:red'><b>0.0061</b></td> </td><td><span style='Color:red'><b>0.9836</b></td> </td><td><span style='Color:red'><b>36.8244</b></td> </td><td><b>52.4920</b></td> </tr> <td>LaBGen </td><td> 5.7214 </td><td> 2.7287 </td><td> 0.7561 </td><td>0.9054 </td><td>29.6545 </td><td>50.7244 </tr> <td>RSL2011 </td><td>7.3911 </td><td>5.3079 </td><td>2.6616 </td><td>0.9310 </td><td>25.3463 </td><td>29.0085 </tr> <td>Photomontage </td><td>6.0480 </td><td>2.4116 </td><td>0.6829 </td><td><b>0.9529</b></td> </td><td>29.4804 </td><td>50.4760 </tr> <td>LRGeomCG </td><td>18.8829 </td><td>35.9543 </td><td>27.5213 </td><td>0.6754 </td><td>20.1325 </td><td>43.7282 </tr> <td>TMac </td><td>19.0408 </td><td>36.2409 </td><td>27.8293 </td><td>0.6726 </td><td>20.0735 </td><td>43.7571</tr> <td>SC-SOBS_1 </td><td><b>4.7184</b></td> </td><td> 4.5671 </td><td><b>0.1616</b></td> </td><td>0.9273 </td><td>29.9489 </td><td><span style='Color:red'><b>54.7757</b></td> </tr> <td>SC-SOBS_2 </td><td>6.5834 </td><td>5.0000 </td><td>0.1829 </td><td>0.8898 </td><td>28.6532 </td><td>51.2832 </tr> <td>BEWIS </td><td>5.5714 </td><td>2.2165 </td><td>0.5274 </td><td>0.9514 </td><td><b>29.9511</b></td> </td><td>50.4528 </tr></tr> <td>Average </td><td>13.4447 </td><td>14.6826 </td><td>10.1115 </td><td>0.8095 </td><td>23.7898 </td><td>42.1286</tr> </table> </span> </div> <br> <div align="center"> <span align="center"> <table align="center" border="1"> <caption>Table 3: Accuracy results of all the compared methods on sequence <i>Candela\_m1.10</i> and their Average. </caption> <tr><td><b>Method</b></td> <td><b>AGE</b></td> <td><b>pEPs</b></td> <td><b>pCEPs</b></td> <td><b>MS-SSIM</b></td> <td><b>PSNR</b></td> <td><b>CQM</b></td></tr> <td>Mean </td><td>3.1270 </td><td>3.0086 </td><td>1.9423 </td><td>0.9503 </td><td>29.4636 </td><td>45.0242 </tr> <td>Color Median </td><td>3.3225 </td><td>3.3519 </td><td>2.1928 </td><td>0.9382 </td><td>27.5054 </td><td>39.9418 </tr> <td>MOG2 </td><td>1.7044 </td><td>0.7694 </td><td>0.6185 </td><td><b>0.9914</b></td> </td><td>34.0895 </td><td>46.9634 </tr> <td>KNN </td><td>11.2176 </td><td> 10.0507 </td><td>5.8179 </td><td>0.8158 </td><td>17.3467 </td><td>23.1109 </tr> <td>BE-AAPSA </td><td>2.2656 </td><td><span style='Color:red'><b>0.0116</b></td> </td><td><span style='Color:red'><b>0.0065</b></td> </td><td>0.9733 </td><td>31.9643 </td><td><b>47.2827</b></td> </tr> <td>WS2006 </td><td>2.5528 </td><td>1.9048 </td><td>0.9657 </td><td>0.9636 </td><td>29.6869 </td><td>40.4489</tr> <td>IMBS-MT </td><td><span style='Color:red'><b>1.3823</b></td> </td><td>0.4705 </td><td><b>0.0957</b></td> </td><td>0.9893 </td><td><b>35.4288</b></td> </td><td>44.2374 </tr> <td>LaBGen </td><td>2.5700 </td><td>1.6809 </td><td>1.2676 </td><td>0.9709 </td><td>30.3140 </td><td>39.7975 </tr> <td>RSL2011 </td><td><b>1.5767</b></td> </td><td><b>0.3748</b></td> </td><td>0.2358 </td><td><span style='Color:red'><b>0.9916</b></td> </td><td><span style='Color:red'><b>36.3572</b></td> </td><td>43.9371 </tr> <td>Photomontage </td><td>3.6780 </td><td>3.5837 </td><td>2.3763 </td><td>0.9332 </td><td>26.8665 </td><td>38.8983 </tr> <td>LRGeomCG </td><td>1.9037 </td><td>0.6579 </td><td>0.4991 </td><td>0.9912 </td><td>33.8805 </td><td>45.0354 </tr> <td>TMac </td><td>2.0456 </td><td>1.0150 </td><td>0.7842 </td><td>0.9888 </td><td>32.5507 </td><td>43.7920</tr> <td>SC-SOBS_1 </td><td>1.8472 </td><td>0.8986 </td><td>0.5080 </td><td>0.9775 </td><td>32.6782 </td><td><span style='Color:red'><b>49.9181</b></td> </tr> <td>SC-SOBS_2 </td><td>3.0125 </td><td>2.2204 </td><td>1.2015 </td><td>0.9532 </td><td>28.9964 </td><td>40.4835 </tr> <td>BEWIS </td><td>1.9049 </td><td>0.7931 </td><td>0.4350 </td><td>0.9852 </td><td>34.0806 </td><td>41.6700 </tr></tr> <td>Average </td><td>2.9407 </td><td>2.0528 </td><td>1.2631 </td><td>0.9609 </td><td>30.7473 </td><td>42.0361 </tr> </table> </span> </div> <br> <div align="center"> <span align="center"> <table align="center" border="1"> <caption>Table 4: Accuracy results of all the compared methods on sequence <i>CAVIAR1</i> and their Average. </caption> <tr><td><b>Method</b></td> <td><b>AGE</b></td> <td><b>pEPs</b></td> <td><b>pCEPs</b></td> <td><b>MS-SSIM</b></td> <td><b>PSNR</b></td> <td><b>CQM</b></td></tr> <td>Mean </td><td>5.2178 </td><td>5.1097 </td><td>4.0141 </td><td>0.9382 </td><td>29.3866 </td><td>51.3920 </tr> <td>Color Median </td><td>2.6858 </td><td>0.3438 </td><td>0.2421 </td><td>0.9918 </td><td>34.8191 </td><td>51.6298 </tr> <td>MOG2 </td><td>3.1274 </td><td>2.8412 </td><td>2.3621 </td><td>0.9722 </td><td>29.3055 </td><td>41.4591 </tr> <td>KNN </td><td>4.6259 </td><td>4.0415 </td><td>2.8463 </td><td>0.9348 </td><td>24.3250 </td><td>29.9878 </tr> <td>BE-AAPSA </td><td>3.6881 </td><td><span style='Color:red'><b>0.0091</b></td> </td><td><span style='Color:red'><b>0.0037</b></td> </td><td>0.9667 </td><td>32.4477 </td><td>51.0556 </tr> <td>WS2006 </td><td>2.6638 </td><td>0.1261 </td><td><b>0.0071</b></td> </td><td>0.9932 </td><td>35.8184 </td><td>49.3176</tr> <td>IMBS-MT </td><td><span style='Color:red'><b>1.2267</b></td> </td><td><b>0.0539</b></td> </td><td>0.0214 </td><td><span style='Color:red'><b>0.9967</b></td> </td><td><span style='Color:red'><b>42.2244</b></td> </td><td><span style='Color:red'><b>55.0816</b></td> </tr> <td>LaBGen </td><td>3.8243 </td><td>0.6327 </td><td>0.4679 </td><td>0.9819 </td><td>31.5534 </td><td>49.2123 </tr> <td>RSL2011 </td><td><b>2.3295</b></td> </td><td>0.1597 </td><td>0.0397 </td><td><b>0.9947</b></td> </td><td><b>37.9348</b></td> </td><td><b>52.3607</b></td> </tr> <td>Photomontage </td><td>2.6498 </td><td>0.1333 </td><td>0.0651 </td><td>0.9933 </td><td>37.1385 </td><td>50.0340 </tr> <td>LRGeomCG </td><td>5.6735 </td><td>6.6274 </td><td>5.4830 </td><td>0.9120 </td><td>28.1790 </td><td>51.1680 </tr> <td>TMac </td><td>5.6945 </td><td>6.7017 </td><td>5.5593 </td><td>0.9116 </td><td>28.1425 </td><td>51.1681</tr> <td>SC-SOBS_1 </td><td>3.0788 </td><td>0.8199 </td><td>0.4944 </td><td>0.9781 </td><td>32.0824 </td><td>51.6212 </tr> <td>SC-SOBS_2 </td><td>4.1143 </td><td>1.0376 </td><td>0.6571 </td><td>0.9724 </td><td>30.2392 </td><td>48.7241 </tr> <td>BEWIS </td><td>3.5539 </td><td>0.4588 </td><td>0.3103 </td><td>0.9813 </td><td>32.2702 </td><td>50.0786 </tr></tr> <td>Average </td><td>3.6103 </td><td>1.9398 </td><td>1.5049 </td><td> 0.9679 </td><td>32.3911 </td><td>48.9527</tr> </table> </span> </div> <br> <div align="center"> <span align="center"> <table align="center" border="1"> <caption>Table 5: Accuracy results of all the compared methods on sequence <i>CAVIAR2</i> and their Average. </caption> <tr><td><b>Method</b></td> <td><b>AGE</b></td> <td><b>pEPs</b></td> <td><b>pCEPs</b></td> <td><b>MS-SSIM</b></td> <td><b>PSNR</b></td> <td><b>CQM</b></td></tr> <td>Mean </td><td>1.1967 </td><td>0.1689 </td><td>0.0356 </td><td>0.9979 </td><td>40.7302 </td><td><span style='Color:red'><b>59.6979</b></td> </tr> <td>Color Median </td><td><span style='Color:red'><b>0.6987</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.9994</b></td> </td><td><b>47.5113</b></td> </td><td><b>59.5439</b></td> </tr> <td>MOG2 </td><td>1.4154 </td><td>0.1658 </td><td>0.0997 </td><td>0.9974 </td><td>40.1120 </td><td>53.4368 </tr> <td>KNN </td><td>7.1935 </td><td>4.8910 </td><td>1.4404 </td><td>0.8469 </td><td>18.9970 </td><td>25.5783 </tr> <td>BE-AAPSA </td><td>1.1718 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9983 </td><td>43.7194 </td><td>54.6637 </tr> <td>WS2006 </td><td>0.7138 </td><td>0.0387 </td><td>0.0000 </td><td>0.9991 </td><td>44.1003 </td><td>59.6514</tr> <td>IMBS-MT </td><td>1.2948 </td><td>0.0102 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9986 </td><td>43.0235 </td><td>53.7161 </tr> <td>LaBGen </td><td>0.8131 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9993 </td><td>46.8425 </td><td>58.3582 </tr> <td>RSL2011 </td><td>0.8678 </td><td>0.1312 </td><td>0.0763 </td><td>0.9962 </td><td>39.7804 </td><td>57.3866 </tr> <td>Photomontage </td><td>1.1047 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9984 </td><td>44.4508 </td><td>54.3645 </tr> <td>LRGeomCG </td><td>1.1822 </td><td>0.3265 </td><td>0.1038 </td><td>0.9971 </td><td>39.8982 </td><td>59.1772 </tr> <td>TMac </td><td>1.1877 </td><td>0.3286 </td><td>0.1058 </td><td>0.9970 </td><td>39.8569 </td><td>59.1736</tr> <td>SC-SOBS_1 </td><td>0.7550 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.9994</b></td> </td><td>47.2190 </td><td>59.1624 </tr> <td>SC-SOBS_2 </td><td>0.9428 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9992 </td><td>45.6705 </td><td>57.4730 </tr> <td>BEWIS </td><td><b>0.7389</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.9994</b></td> </td><td><span style='Color:red'><b>47.6100</b></td> </td><td>58.7359 </tr></tr> <td>Average </td><td>1.4185 </td><td>0.4041 </td><td>0.1241 </td><td>0.9882 </td><td>41.9681 </td><td>55.3413</tr> </table> </span> </div> <br> <div align="center"> <span align="center"> <table align="center" border="1"> <caption>Table 6: Accuracy results of all the compared methods on sequence <i>CaVignal</i> and their Average. </caption> <tr><td><b>Method</b></td> <td><b>AGE</b></td> <td><b>pEPs</b></td> <td><b>pCEPs</b></td> <td><b>MS-SSIM</b></td> <td><b>PSNR</b></td> <td><b>CQM</b></td></tr> <td>Mean </td><td>8.7869 </td><td>10.4890 </td><td>8.2537 </td><td>0.8338 </td><td>21.6405 </td><td>50.2703 </tr> <td>Color Median </td><td>10.3082 </td><td>10.4632 </td><td>8.1066 </td><td>0.7984 </td><td>18.1355 </td><td>33.1438 </tr> <td>MOG2 </td><td>16.9327 </td><td>0.1114 </td><td>0.0837 </td><td>0.8136 </td><td>18.5891 </td><td>34.5104 </tr> <td>KNN </td><td>15.9267 </td><td>0.0813 </td><td>0.0127 </td><td>0.8241 </td><td>18.2332 </td><td>30.9930 </tr> <td>BE-AAPSA </td><td>10.0755 </td><td>4.8100 </td><td>3.1200 </td><td>0.9711 </td><td>26.1972 </td><td>39.4600</tr> <td>WS2006 </td><td>2.5403 </td><td>1.5000 </td><td>0.4743 </td><td>0.9289 </td><td>27.1089 </td><td>37.0609</tr> <td>IMBS-MT </td><td>0.7692</td><td><span style='Color:red'><b>0.0147</b></td> </td><td><span style='Color:red'><b>0.0000</b></td></td><td><b>0.9982</b></td> </td><td><b>45.9202</b></td> </td><td><b>57.1044</b></td> </tr> <td>LaBGen </td><td><b>0.4542</b></td> </td><td><span style='Color:red'><b>0.0147</b></td> </td><td><span style='Color:red'><b>0.0000</b></td></td><td>0.9981 </td><td>45.5789 </td><td>55.9161 </tr> <td>RSL2011 </td><td>0.9106 </td><td><span style='Color:red'><b>0.0147</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9973 </td><td>43.9322 </td><td>53.7718 </tr> <td>Photomontage </td><td>11.2665 </td><td>11.2206 </td><td>8.8529 </td><td>0.7919 </td><td>17.6257 </td><td>32.0570 </tr> <td>LRGeomCG </td><td>5.4839 </td><td>6.4118 </td><td>3.9890 </td><td>0.9111 </td><td>28.3288 </td><td>52.9853 </tr> <td>TMac </td><td>5.4979 </td><td>6.5074 </td><td>4.0625 </td><td>0.9109 </td><td>28.2877 </td><td>53.0188</tr> <td>SC-SOBS_1 </td><td>0.9590 </td><td>0.6728 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9947 </td><td>37.4679 </td><td>55.9939 </tr> <td>SC-SOBS_2 </td><td>1.1434 </td><td>0.6949 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9935 </td><td>37.0992 </td><td>54.1992</tr> <td>BEWIS </td><td><span style='Color:red'><b>0.3990</b></td> </td><td><span style='Color:red'><b>0.0147</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.9984</b></td> </td><td><span style='Color:red'><b>46.5616</b></td> </td><td><span style='Color:red'><b>59.8510</b></td></tr></tr> <td>Average </td><td>6.0969 </td><td>3.5347 </td><td>2.4637 </td><td>0.9176 </td><td>30.7138 </td><td>46.6891</tr> </table> </span> </div> <br> <div align="center"> <span align="center"> <table align="center" border="1"> <caption>Table 7: Accuracy results of all the compared methods on sequence <i>Foliage</i> and their Average. </caption> <tr><td><b>Method</b></td> <td><b>AGE</b></td> <td><b>pEPs</b></td> <td><b>pCEPs</b></td> <td><b>MS-SSIM</b></td> <td><b>PSNR</b></td> <td><b>CQM</b></td></tr> <td>Mean </td><td>30.0992 </td><td>65.8542 </td><td>46.9549 </td><td>0.6785 </td><td>17.3201 </td><td>26.8996 </tr> <td>Color Median </td><td>27.0135 </td><td>47.3125 </td><td>30.4583 </td><td>0.6444 </td><td>16.7842 </td><td>28.7321 </tr> <td>MOG2 </td><td>32.3624 </td><td>0.6685 </td><td>0.5526 </td><td>0.8038 </td><td>16.5991 </td><td>31.5282 </tr> <td>KNN </td><td>34.5615 </td><td>0.3962 </td><td>0.0385 </td><td>0.6281 </td><td>14.1761 </td><td>25.6845 </tr> <td>BE-AAPSA </td><td>26.2190 </td><td>59.9800 </td><td>43.0900 </td><td>0.8015 </td><td>18.4317 </td><td>30.2999 </tr> <td>WS2006 </td><td>6.8649 </td><td>2.8507 </td><td>0.0069 </td><td>0.9754 </td><td>27.2438 </td><td>34.9776</tr> <td>IMBS-MT </td><td>7.5809 </td><td>9.8507 </td><td>3.1319 </td><td>0.9090 </td><td>22.7278 </td><td>34.0028 </tr> <td>LaBGen </td><td><span style='Color:red'><b>1.6172</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.9982</b></td> </td><td><span style='Color:red'><b>40.6051</b></td> </td><td><span style='Color:red'><b>46.7662</b></td> </tr> <td>RSL2011 </td><td>9.0230 </td><td>12.3090 </td><td>8.1250 </td><td>0.8370 </td><td>20.9844 </td><td>30.4461 </tr> <td>Photomontage </td><td>1.8592 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9974 </td><td>39.1779 </td><td>45.6052 </tr> <td>LRGeomCG </td><td>11.6932 </td><td>20.8924 </td><td>14.9861 </td><td>0.9535 </td><td>23.9826 </td><td>39.0643 </tr> <td>TMac </td><td>12.0335 </td><td>21.9826 </td><td>15.9861 </td><td>0.9498 </td><td>23.7072 </td><td>38.7549</tr> <td>SC-SOBS_1 </td><td>3.0825 </td><td>0.0625 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9939 </td><td>35.6936 </td><td>39.6256 </tr> <td>SC-SOBS_2 </td><td>3.3587 </td><td>0.0660 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9931 </td><td>35.1103 </td><td>39.5048 </tr> <td>BEWIS </td><td><b>1.7767</b></td> </td><td><b>0.0174</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><b>0.9978</b></td> </td><td><b>39.5441</b></td> </td><td><b>46.1713</b></td> </tr></tr> <td>Average </td><td>13.9430 </td><td>16.1495 </td><td>10.8887 </td><td>0.8774 </td><td>26.1392 </td><td>35.8709</tr> </table> </span> </div> <br> <div align="center"> <span align="center"> <table align="center" border="1"> <caption>Table 8: Accuracy results of all the compared methods on sequence <i>Hall&Monitor</i> and their Average. </caption> <tr><td><b>Method</b></td> <td><b>AGE</b></td> <td><b>pEPs</b></td> <td><b>pCEPs</b></td> <td><b>MS-SSIM</b></td> <td><b>PSNR</b></td> <td><b>CQM</b></td></tr> <td>Mean </td><td>3.7238 </td><td>3.0516 </td><td>1.6643 </td><td>0.9545 </td><td>29.7571 </td><td>42.6596 </tr> <td>Color Median </td><td>2.7105 </td><td>0.9931 </td><td>0.5339 </td><td>0.9640 </td><td>30.4656 </td><td>42.6705 </tr> <td>MOG2 </td><td>2.4506 </td><td><span style='Color:red'><b>0.0109</b></td> </td><td>0.0045 </td><td>0.9833 </td><td>34.3943 </td><td>45.9714 </tr> <td>KNN </td><td>3.9413 </td><td><b>0.0121</b></td> </td><td>0.0021 </td><td>0.9519 </td><td>28.2208 </td><td>37.4907 </tr> <td>BE-AAPSA </td><td>2.4425 </td><td>0.3200 </td><td>0.0400 </td><td>0.9892 </td><td>36.4218 </td><td>45.2466 </tr> <td>WS2006 </td><td>2.6644 </td><td>0.5563 </td><td>0.0308 </td><td>0.9821 </td><td>30.9313 </td><td>40.0949</tr> <td>IMBS-MT </td><td><span style='Color:red'><b>1.5350</b></td> </td><td>0.0923 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.9954</b></td> </td><td><span style='Color:red'><b>38.6214</b></td> </td><td><span style='Color:red'><b>48.5224</b></td> </tr> <td>LaBGen </td><td>2.4008 </td><td>0.1302 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9916 </td><td>37.1746 </td><td>45.1416 </tr> <td>RSL2011 </td><td>3.2937 </td><td>1.6489 </td><td>0.7931 </td><td>0.9377 </td><td>26.9214 </td><td>36.7046 </tr> <td>Photomontage </td><td>2.7986 </td><td>0.3610 </td><td>0.0817 </td><td>0.9819 </td><td>33.3715 </td><td>41.7323 </tr> <td>LRGeomCG </td><td>2.0476 </td><td>0.2237 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><b>0.9938</b></td> </td><td><b>38.0243</b></td> </td><td><b>46.3813</b></td> </tr> <td>TMac </td><td>2.0599 </td><td>0.2415 </td><td> <span style='Color:red'><b>0.0000</b></td> </td><td>0.9937 </td><td>37.5664 </td><td>46.2214</tr> <td>SC-SOBS_1 </td><td><b>1.8125</b></td> </td><td>0.6641 </td><td>0.2166 </td><td>0.9832 </td><td>34.2985 </td><td>44.2863 </tr> <td>SC-SOBS_2 </td><td>2.6930 </td><td>0.7599 </td><td>0.2166 </td><td>0.9798 </td><td>33.1795 </td><td>42.5386 </tr> <td>BEWIS </td><td>3.6217 </td><td>1.4347 </td><td>0.0154 </td><td>0.9626 </td><td>27.1794 </td><td>35.6121 </tr></tr> <td>Average </td><td>2.6797 </td><td>0.7000 </td><td>0.2399 </td><td>0.9763 </td><td>33.1019 </td><td>42.7516</tr> </table> </span> </div> <br> <div align="center"> <span align="center"> <table align="center" border="1"> <caption>Table 9: Accuracy results of all the compared methods on sequence <i>HighwayI</i> and their Average. </caption> <tr><td><b>Method</b></td> <td><b>AGE</b></td> <td><b>pEPs</b></td> <td><b>pCEPs</b></td> <td><b>MS-SSIM</b></td> <td><b>PSNR</b></td> <td><b>CQM</b></td></tr> <td>Mean </td><td>6.4127 </td><td>0.2995 </td><td>0.0156 </td><td>0.9700 </td><td>30.3580 </td><td>60.3090 </tr> <td>Color Median </td><td><b>1.4275</b></td> </td><td>0.1563 </td><td>0.0143 </td><td>0.9924 </td><td>40.1432 </td><td><b>62.5723</b></td> </tr> <td>MOG2 </td><td>2.6031 </td><td><span style='Color:red'><b>0.0023</b></td> </td><td><b>0.0002</b></td> </td><td>0.9753 </td><td>35.8635 </td><td>58.2889 </tr> <td>KNN </td><td>6.1277 </td><td>0.0616 </td><td>0.0003 </td><td>0.8506 </td><td>25.1521 </td><td>34.8174 </tr> <td>BE-AAPSA </td><td>4.3721 </td><td>2.7600 </td><td>0.6900 </td><td>0.9442 </td><td>31.1332 </td><td>52.3623 </tr> <td>WS2006 </td><td>2.5185 </td><td>0.6849 </td><td>0.0247 </td><td>0.9816 </td><td>35.6885 </td><td>56.9113</tr> <td>IMBS-MT </td><td>1.4913 </td><td>0.0612 </td><td>0.0026 </td><td><b>0.9939</b></td> </td><td><b>41.7728</b></td> </td><td>58.8328 </tr> <td>LaBGen </td><td>1.9054 </td><td>0.4362 </td><td>0.0286 </td><td>0.9877 </td><td>37.4928 </td><td>53.1613 </tr> <td>RSL2011 </td><td>1.5918 </td><td>0.2344 </td><td>0.0195 </td><td>0.9899 </td><td>38.8728 </td><td>59.4531 </tr> <td>Photomontage </td><td>2.1745 </td><td>0.4076 </td><td>0.0482 </td><td>0.9830 </td><td>37.1250 </td><td>59.0270 </tr> <td>LRGeomCG </td><td>2.6535 </td><td>0.2018 </td><td>0.0130 </td><td>0.9779 </td><td>36.2808 </td><td>58.2359 </tr> <td>TMac </td><td>2.6788 </td><td>0.1992 </td><td>0.0130 </td><td>0.9777 </td><td>36.1917 </td><td>58.2417</tr> <td>SC-SOBS_1 </td><td><span style='Color:red'><b>0.9917</b></td> </td><td><b>0.0026</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.9968</b></td> </td><td><span style='Color:red'><b>44.3343</b></td> </td><td><span style='Color:red'><b>66.0819 </b></td></tr> <td>SC-SOBS_2 </td><td>2.1209 </td><td>0.3216 </td><td>0.0221 </td><td>0.9870 </td><td>37.3789 </td><td>53.6069 </tr> <td>BEWIS </td><td>2.1070 </td><td>0.4661 </td><td>0.0169 </td><td>0.9886 </td><td>36.8023 </td><td>54.4956 </tr></tr> <td>Average </td><td>2.7451 </td><td>0.4197 </td><td>0.0606 </td><td>0.9731 </td><td>36.3060 </td><td>56.4265</tr> </table> </span> </div> <br> <div align="center"> <span align="center"> <table align="center" border="1"> <caption>Table 10: Accuracy results of all the compared methods on sequence <i>HighwayII</i> and their Average. </caption> <tr><td><b>Method</b></td> <td><b>AGE</b></td> <td><b>pEPs</b></td> <td><b>pCEPs</b></td> <td><b>MS-SSIM</b></td> <td><b>PSNR</b></td> <td><b>CQM</b></td></tr> <td>Mean </td><td>3.3414 </td><td>0.3607 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9915 </td><td>34.3045 </td><td>47.1299 </tr> <td>Color Median </td><td>1.7278 </td><td>0.3190 </td><td>0.0013 </td><td><b>0.9961</b></td> </td><td>34.6639 </td><td>42.3162 </tr> <td>MOG2 </td><td>2.0893 </td><td><b>0.0040</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9946 </td><td>36.1190 </td><td>45.2643 </tr> <td>KNN </td><td>3.2112 </td><td>0.0085 </td><td>0.0001 </td><td>0.9851 </td><td>32.0981 </td><td>39.6454 </tr> <td>BE-AAPSA </td><td>2.5181 </td><td>0.2800 </td><td>0.0100 </td><td>0.9903 </td><td>36.2738 </td><td>47.3613 </tr> <td>WS2006 </td><td>2.4906 </td><td>0.4883 </td><td>0.0130 </td><td>0.9927 </td><td>33.9515 </td><td>40.5088</tr> <td>IMBS-MT </td><td><b>1.8684</b></td> </td><td>0.0260 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9960 </td><td><b>40.1098</b></td> </td><td><b>48.8094</b></td> </tr> <td>LaBGen </td><td>2.4240 </td><td>0.3034 </td><td>0.0026 </td><td>0.9921 </td><td>35.5876 </td><td>42.8025 </tr> <td>RSL2011 </td><td>2.3000 </td><td>0.5130 </td><td>0.0846 </td><td>0.9907 </td><td>33.8305 </td><td>42.4937 </tr> <td>Photomontage </td><td>2.4306 </td><td>0.5885 </td><td>0.0052 </td><td>0.9909 </td><td>34.3975 </td><td>41.7656 </tr> <td>LRGeomCG </td><td>2.7526 </td><td>0.3555 </td><td>0.0026 </td><td>0.9908 </td><td>35.3406 </td><td>46.3161 </tr> <td>TMac </td><td>2.7697 </td><td>0.3763 </td><td> 0.0039 </td><td> 0.9908 </td><td>35.2287 </td><td>46.2181</tr> <td>SC-SOBS_1 </td><td><span style='Color:red'><b>0.7100</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.9991</b></td> </td><td><span style='Color:red'><b>46.8739</b></td> </td><td><span style='Color:red'><b>56.6012</b></td> </tr> <td>SC-SOBS_2 </td><td>2.3946 </td><td>0.2982 </td><td>0.0039 </td><td>0.9926 </td><td>35.7688 </td><td>43.2384 </tr> <td>BEWIS </td><td>2.1932 </td><td>0.4141 </td><td>0.0013 </td><td>0.9942 </td><td>34.6264 </td><td>41.4061 </tr></tr> <td>Average </td><td>2.3481 </td><td>0.2890 </td><td>0.0086 </td><td>0.9925 </td><td>35.9450 </td><td>44.7918 </tr> </table> </span> </div> <br> <div align="center"> <span align="center"> <table align="center" border="1"> <caption>Table 11: Accuracy results of all the compared methods on sequence <i>HumanBody2</i> and their Average. </caption> <tr><td><b>Method</b></td> <td><b>AGE</b></td> <td><b>pEPs</b></td> <td><b>pCEPs</b></td> <td><b>MS-SSIM</b></td> <td><b>PSNR</b></td> <td><b>CQM</b></td></tr> <td>Mean </td><td>6.3783 </td><td>5.5065 </td><td>3.0443 </td><td>0.9736 </td><td>27.9022 </td><td>38.4071 </tr> <td>Color Median </td><td>2.9408 </td><td>0.2995 </td><td>0.0391 </td><td>0.9970 </td><td>35.4279 </td><td>46.6252 </tr> <td>MOG2 </td><td>11.2767 </td><td>13.4609 </td><td>9.9427 </td><td>0.8752 </td><td>19.5258 </td><td>32.1251 </tr> <td>KNN </td><td>20.9423 </td><td>18.5130 </td><td>15.2188 </td><td>0.7783 </td><td>14.5871 </td><td>21.4805 </tr> <td>BE-AAPSA </td><td>6.3274 </td><td><span style='Color:red'><b>0.0797</b></td> </td><td>0.0550 </td><td>0.9528 </td><td>24.9434 </td><td>36.6271 </tr> <td>WS2006 </td><td>3.9876 </td><td>0.6393 </td><td>0.0026 </td><td>0.9923 </td><td>30.7994 </td><td>42.8966</tr> <td>IMBS-MT </td><td><b>1.9190</b></td> </td><td>0.5794 </td><td>0.0534 </td><td>0.9958 </td><td>34.0997 </td><td>45.2074 </tr> <td>LaBGen </td><td>3.8273 </td><td>0.2630 </td><td>0.0013 </td><td><b>0.9975</b></td> </td><td>34.4291 </td><td><b>46.8653</b></td> </tr> <td>RSL2011 </td><td>3.1154 </td><td>0.3099 </td><td>0.0013 </td><td>0.9959 </td><td><b>35.5261</b></td> </td><td>46.2671 </tr> <td>Photomontage </td><td>11.4203 </td><td>13.0052 </td><td>9.4375 </td><td>0.8751 </td><td>19.2008 </td><td>30.9881 </tr> <td>LRGeomCG </td><td>5.7621 </td><td>4.6497 </td><td>2.4414 </td><td>0.9788 </td><td>28.7047 </td><td>40.1984 </tr> <td>TMac </td><td>5.8044 </td><td>4.7292 </td><td>2.4583 </td><td>0.9786 </td><td>28.6019 </td><td>40.2122</tr> <td>SC-SOBS_1 </td><td><span style='Color:red'><b>1.8126</b></td> </td><td><b>0.0990</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.9980</b></td> </td><td><span style='Color:red'><b>39.3952</b></td> </td><td><span style='Color:red'><b>47.3105</b></td> </tr> <td>SC-SOBS_2 </td><td>3.3927 </td><td>0.3411 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9969 </td><td>35.0465 </td><td>45.4292 </tr> <td>BEWIS </td><td>4.2667 </td><td>1.5013 </td><td>0.0260 </td><td>0.9866 </td><td>27.9740 </td><td>41.7024 </tr></tr> <td>Average </td><td>6.2116 </td><td>4.2651 </td><td>2.8481 </td><td>0.9582 </td><td>29.0776 </td><td>40.1561 </tr> </table> </span> </div> <br> <div align="center"> <span align="center"> <table align="center" border="1"> <caption>Table 12: Accuracy results of all the compared methods on sequence <i>IBMtest2</i> and their Average. </caption> <tr><td><b>Method</b></td> <td><b>AGE</b></td> <td><b>pEPs</b></td> <td><b>pCEPs</b></td> <td><b>MS-SSIM</b></td> <td><b>PSNR</b></td> <td><b>CQM</b></td></tr> <td>Mean </td><td>5.1227 </td><td>3.7643 </td><td>1.8112 </td><td>0.9800 </td><td>30.2187 </td><td>41.6366 </tr> <td>Color Median </td><td><span style='Color:red'><b>2.2862</b></td> </td><td>0.0391 </td><td><span style='Color:red'><b>0.0000</b></td></td><td><b>0.9939</b></td> </td><td><b>36.6967</b></td> </td><td><b>48.5607</b></td> </tr> <td>MOG2 </td><td>3.1981 </td><td>1.5039 </td><td>0.7083 </td><td>0.9717 </td><td>30.518 </td><td>38.1524 </tr> <td>KNN </td><td>21.3572 </td><td>16.2995 </td><td>2.3099 </td><td>0.6671 </td><td>14.1235 </td><td>20.5705 </tr> <td>BE-AAPSA </td><td>5.7290 </td><td><span style='Color:red'><b>0.0012</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9914 </td><td>31.7541 </td><td>44.5344 </tr> <td>WS2006 </td><td>4.6744 </td><td>1.9531 </td><td>0.0495 </td><td>0.9410 </td><td>24.2631 </td><td>32.8595</tr> <td>IMBS-MT </td><td>7.3508 </td><td>3.2734 </td><td>0.1328 </td><td>0.9721 </td><td>24.6275 </td><td>36.4310 </tr> <td>LaBGen </td><td>3.7491 </td><td>0.0872 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9906 </td><td>33.5923 </td><td>45.9029 </tr> <td>RSL2011 </td><td>6.1074 </td><td>2.7005 </td><td>0.9922 </td><td>0.9303 </td><td>24.4272 </td><td>36.0603 </tr> <td>Photomontage </td><td>3.1954 </td><td>0.0690 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9898 </td><td>35.1813 </td><td>45.3905 </tr> <td>LRGeomCG </td><td>3.6413 </td><td>1.4544 </td><td>0.6081 </td><td>0.9868 </td><td>32.8930 </td><td>44.9516 </tr> <td>TMac </td><td>3.6575 </td><td>1.4831 </td><td>0.6289 </td><td>0.9868 </td><td>32.8424 </td><td>44.9660</tr> <td>SC-SOBS_1 </td><td><b>2.3424</b></td> </td><td><b>0.0143</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.9954</b></td> </td><td><span style='Color:red'><b>37.9515</b></td> </td><td><span style='Color:red'><b>50.5923</b></td> </tr> <td>SC-SOBS_2 </td><td>3.9729 </td><td>0.0964 </td><td><span style='Color:red'><b>0.0000</b></td> </td><td>0.9919 </td><td>33.7736 </td><td>46.3316 </tr> <td>BEWIS </td><td>3.9848 </td><td>1.5013 </td><td>0.0651 </td><td>0.9602 </td><td>25.6501 </td><td>39.8792 </tr></tr> <td>Average </td><td>5.3579 </td><td>2.2827 </td><td>0.4871 </td><td>0.9566</td><td>29.9009 </td><td>41.1213 </tr> </table> </span> </div> <br> <div align="center"> <span align="center"> <table align="center" border="1"> <caption>Table 13: Accuracy results of all the compared methods on sequence <i>People&Foliage</i> and their Average. </caption> <tr><td><b>Method</b></td> <td><b>AGE</b></td> <td><b>pEPs</b></td> <td><b>pCEPs</b></td> <td><b>MS-SSIM</b></td> <td><b>PSNR</b></td> <td><b>CQM</b></td></tr> <td>Mean </td><td>34.4507 </td><td>61.6576 </td><td>54.0924 </td><td>0.7555 </td><td>15.3719 </td><td>31.9727 </tr> <td>Color Median </td><td>24.4211 </td><td>32.2396 </td><td>25.3203 </td><td>0.6114 </td><td>15.1870 </td><td>27.4979 </tr> <td>MOG2 </td><td>33.8442 </td><td>0.7108 </td><td>0.6134 </td><td>0.8584 </td><td>16.2252 </td><td>27.4728 </tr> <td>KNN </td><td>48.4920 </td><td>0.4718 </td><td>0.2966 </td><td>0.4238 </td><td>10.9196 </td><td>19.8121 </tr> <td>BE-AAPSA </td><td>20.1865 </td><td>31.0000 </td><td>24.1900 </td><td>0.9256 </td><td>19.7152 </td><td>29.3564 </tr> <td>WS2006 </td><td>5.4243 </td><td>3.5716 </td><td>0.0924 </td><td>0.9269 </td><td>22.6952 </td><td>31.3847</tr> <td>IMBS-MT </td><td>8.3982 </td><td>7.3568 </td><td>3.2305 </td><td>0.8514 </td><td>20.0658 </td><td>32.5231 </tr> <td>LaBGen </td><td><b>1.7751</b></td> </td><td><span style='Color:red'><b>0.0026</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><b>0.9968</b></td> </td><td><b>39.7161</b></td> </td><td><b>46.2148</b></td> </tr> <td>RSL2011 </td><td>8.1966 </td><td>9.4023 </td><td>7.8867 </td><td>0.8628 </td><td>21.2093 </td><td>27.1459 </tr> <td>Photomontage </td><td><span style='Color:red'><b>1.4103</b></td> </td><td><b>0.0039</b></td> </td><td><span style='Color:red'><b>0.0000</b></td> </td><td><span style='Color:red'><b>0.9973</b></td> </td><td><span style='Color:red'><b>41.0866</b></td> </td><td><span style='Color:red'><b>47.1517 </b></td></tr> <td>LRGeomCG </td><td>29.2393 </td><td>57.7812 </td><td>48.2135 </td><td>0.8332 </td><td>16.8399 </td><td>32.1823 </tr> <td>TMac </td><td>29.4402 </td><td>57.1888 </td><td>47.6719 </td><td>0.8233 </td><td>16.6676 </td><td>32.1267</tr> <td>SC-SOBS_1 </td><td>7.5889 </td><td>11.5482 </td><td>6.2734 </td><td>0.9333 </td><td>23.6413 </td><td>36.3079 </tr> <td>SC-SOBS_2 </td><td>7.9408 </td><td>11.5586 </td><td>6.2734 </td><td>0.9329 </td><td>23.6085 </td><td>36.1581 </tr> <td>BEWIS </td><td>11.9685 </td><td>13.0182 </td><td>10.2018 </td><td>0.8823 </td><td>17.6743 </td><td>26.7312 </tr></tr> <td>Average </td><td>18.1851 </td><td>19.8341 </td><td>15.6238 </td><td>0.8410 </td><td>21.3749 </td><td>32.2692</tr> </table> </span> </div> <br> <div align="center"> <span align="center"> <table align="center" border="1"> <caption>Table 14: Accuracy results of all the compared methods on sequence <i>Snellen</i> and their Average. </caption> <tr><td><b>Method</b></td> <td><b>AGE</b></td> <td><b>pEPs</b></td> <td><b>pCEPs</b></td> <td><b>MS-SSIM</b></td> <td><b>PSNR</b></td> <td><b>CQM</b></td></tr> <td>Mean </td><td>54.2865 </td><td>87.7025 </td><td>81.0282 </td><td>0.7154 </td><td>12.6049 </td><td>34.4532 </tr> <td>Color Median </td><td>42.3981 </td><td>62.2010 </td><td>56.9734 </td><td>0.6932 </td><td>13.6573 </td><td>36.0691 </tr> <td>MOG2 </td><td>58.8159 </td><td><b>0.7615</b></td> </td><td><b>0.6839</b></td> </td><td>0.5336 </td><td>11.4143 </td><td>27.0312 </tr> <td>KNN </td><td>61.9389 </td><td><span style='Color:red'><b>0.6832</b></td> </td><td><span style='Color:red'><b>0.4328</b></td> </td><td>0.4493 </td><td>10.6164 </td><td>22.5804 </tr> <td>BE-AAPSA </td><td>46.7580 </td><td>76.0800 </td><td>69.3300 </td><td>0.7615 </td><td>13.6310 </td><td>37.1410 </tr> <td>WS2006 </td><td>23.0010 </td><td>23.1674 </td><td>12.2685 </td><td>0.7481 </td><td>15.6158 </td><td>24.9930</tr> <td>IMBS-MT </td><td>14.4480 </td><td>25.3279 </td><td>19.7290 </td><td>0.8668 </td><td>19.7436 </td><td>40.1151 </tr> <td>LaBGen </td><td><b>4.6412</b></td> </td><td>6.3368 </td><td>5.9317 </td><td><span style='Color:red'><b>0.9792</b></td> </td><td><span style='Color:red'><b>27.1445</b></td> </td><td><span style='Color:red'><b>47.9560</b></td> </tr> <td>RSL2011 </td><td>16.0515 </td><td>14.4290 </td><td>12.6640 </td><td>0.7190 </td><td>16.7070 </td><td>28.4869 </tr> <td>Photomontage </td><td>29.9797 </td><td>33.4973 </td><td>30.4688 </td><td>0.5926 </td><td>14.1466 </td><td>26.9210 </tr> <td>LRGeomCG </td><td>24.4846 </td><td>50.4340 </td><td>42.8337 </td><td>0.9250 </td><td>18.6585 </td><td>42.1307 </tr> <td>TMac </td><td>24.8743 </td><td>51.9965 </td><td>44.3528 </td><td>0.9206 </td><td>18.5311 </td><td>41.8552</tr> <td>SC-SOBS_1 </td><td>16.1433 </td><td>35.4504 </td><td>21.8412 </td><td>0.9332 </td><td>21.6050 </td><td><b>46.0165</b></td> </tr> <td>SC-SOBS_2 </td><td>16.5042 </td><td>35.4745 </td><td>21.9088 </td><td>0.9322 </td><td>21.3953 </td><td>44.9320 </tr> <td>BEWIS </td><td><span style='Color:red'><b>4.6386</b></td> </td><td>5.2758 </td><td>3.3131 </td><td><b>0.9692</b></td> </td><td><b>25.7540</b></td> </td><td>42.7116 </tr></tr> <td>Average </td><td>29.2643 </td><td>33.9212 </td><td>28.2507 </td><td>0.7826 </td><td>17.4150 </td><td>36.2262</tr> </table> </span> </div> <br> <div align="center"> <span align="center"> <table align="center" border="1"> <caption>Table 15: Accuracy results of all the compared methods on sequence <i>Toscana</i> and their Average. </caption> <tr><td><b>Method</b></td> <td><b>AGE</b></td> <td><b>pEPs</b></td> <td><b>pCEPs</b></td> <td><b>MS-SSIM</b></td> <td><b>PSNR</b></td> <td><b>CQM</b></td></tr> <td>Mean </td><td>11.6247 </td><td>22.8308 </td><td>17.5896 </td><td>0.8831 </td><td>22.7298 </td><td>36.2525 </tr> <td>Color Median </td><td>5.3148 </td><td>6.4562 </td><td>3.7742 </td><td>0.8707 </td><td>23.2941 </td><td>31.7804 </tr> <td>MOG2 </td><td>9.5929 </td><td>12.806 </td><td>8.3773 </td><td>0.8947 </td><td>23.5968 </td><td>23.1972 </tr> <td>KNN </td><td>19.0935 </td><td>22.7581 </td><td>17.5800 </td><td>0.7034 </td><td>16.3492 </td><td>16.2754</tr> <td>BE-AAPSA </td><td>7.3553 </td><td><span style='Color:red'><b>0.1033</b></td> </td><td><span style='Color:red'><b>0.0658</b></td> </td><td>0.9095 </td><td>25.6164 </td><td>37.6754 </tr> <td>WS2006 </td><td>5.8222 </td><td>5.8935 </td><td>2.3500 </td><td>0.8623 </td><td>22.8504 </td><td>29.0927</tr> <td>IMBS-MT </td><td>7.4109 </td><td>6.9096 </td><td>5.2394 </td><td>0.8903 </td><td>22.5367 </td><td>22.0319 </tr> <td>LaBGen </td><td>6.1993 </td><td>6.9440 </td><td>4.4881 </td><td>0.8805 </td><td>23.1537 </td><td>33.3061 </tr> <td>RSL2011 </td><td>18.7636 </td><td>27.3794 </td><td>22.6806 </td><td>0.6662 </td><td>17.1506 </td><td>24.4755 </tr> <td>Photomontage </td><td><span style='Color:red'><b>1.5175</b></td> </td><td><b>0.4517</b></td> </td><td><b>0.1652</b></td> </td><td><span style='Color:red'><b>0.9892</b></td> </td><td><span style='Color:red'><b>36.7526</b></td> </td><td><span style='Color:red'><b>50.2416</b></td> </tr> <td>LRGeomCG </td><td>7.3009 </td><td>11.8569 </td><td>8.4394 </td><td>0.8959 </td><td>24.2914 </td><td>36.3206 </tr> <td>TMac </td><td>7.3742 </td><td>12.0163 </td><td>8.5815 </td><td>0.8958 </td><td>24.2609 </td><td>36.2690</tr> <td>SC-SOBS_1 </td><td><b>3.1898</b></td> </td><td>3.3113 </td><td>1.7183 </td><td><b>0.9616</b></td> </td><td><b>30.6221</b></td> </td><td><b>43.3002</b></td> </tr> <td>SC-SOBS_2 </td><td>6.2949 </td><td>8.5400 </td><td>5.0521 </td><td>0.8880 </td><td>24.9143 </td><td>36.7567 </tr> <td>BEWIS </td><td>7.4054 </td><td>6.8877 </td><td>5.0215 </td><td>0.8878 </td><td>22.5227 </td><td>31.7212 </tr></tr> <td>Average </td><td>8.2840 </td><td>10.3430 </td><td>7.4082 </td><td>0.8719 </td><td>24.0428 </td><td>32.5798 </tr> </table> </span> </div> <br> Please, observe that the above CMQ values were evaluated using a previous version of the Matlab scripts, that included a bug (you can still download the <a href=./MODLab/BckgInit/MATLAB/EvaluateBckgInit3Old.zip>Old version of Matlab scripts</a> to compare with the above results). <p><h3>References</h3></p> <p><a name=[1]><b><span style='color:blue'>[1]</span></b> L. Maddalena, A. Petrosino, <a target="_blank" href=http://link.springer.com/chapter/10.1007/978-3-319-23222-5_57#> <i>Towards Benchmarking Scene Background Initialization</i></a>, in V. Murino et al. (eds), <a target="_blank" href=http://www.springer.com/us/book/9783319232218>New Trends in Image Analysis and Processing -- ICIAP 2015 Workshops</a>, Lecture Notes in Computer Science, Vol. 9281, Springer International Publishing Switzerland, DOI 10.1007/978-3-319-23222-5_57#, pp. 469 476, 2015. <p><a name=[2]><b><span style='color:blue'>[2]</span></b> T. Bouwmans, L. Maddalena, A. Petrosino, <a target="_blank" href=https://www.sciencedirect.com/science/article/abs/pii/S0167865516303798><i>Scene background initialization: A taxonomy</i></a>, Pattern Recognition Letters 96, DOI 10.1016/j.patrec.2016.12.024, 3-11, 2017.</p> </span> </span> </body> </html>