Using the AI capabilities of Google Cloud, CyberLink enables visual content creators to deliver appealing work more efficiently.
Using the AI capabilities of Google Cloud, CyberLink enables visual content creators to deliver appealing work more efficiently.
Empowering visual content creation with AI
The rise of generative AI is quickly changing how visual content is created. CyberLink, a multimedia software provider with nearly 30 years of experience, sees this as an opportunity to help people push the boundaries of their creativity.
“Generative AI is becoming an integral part of multimedia creation,” says Vincent Lin, president of Cyberlink. “We believe that on top of assisting people to realize their creative ideas more efficiently, AI can play a collaborative role and provide boundless inspiration.”
CyberLink has been developing AI solutions for nearly a decade. Its video and photo editing applications contain a variety of AI features, and its portfolio includes AI-powered facial recognition.
Google Cloud provides not only generative AI models with outstanding comprehension capability and output quality, but also a wide range of AI solutions that can be seamlessly integrated for more complex use cases.
Vincent Lin
President, CyberLink
But as generative AI took off, so did the demand for compute. That’s when CyberLink realized it needed external resources to bring the latest technologies faster to its users.
To keep pace and bring the latest tech to market faster, the company selected Google Cloud because it offers diverse models with good compatibility, while Gemini fulfils CyberLink’s performance requirements for video editing.
“To enable our users to fully enjoy the advantages of AI technology advancement, we need to collaborate with technology leaders in the field,” explains Lin. “Google Cloud provides not only generative AI models with outstanding comprehension capability and output quality, but also a wide range of AI solutions that can be seamlessly integrated for more complex use cases.”
Enabling efficient, creative visual content editing with Gemini and Imagen
In mid-2025, CyberLink launched a new auto-edit feature in its video editing app. Powered by Gemini 2.5 Flash Lite, it turns raw footage imported by users into polished short videos, complete with captions and background music, in under one minute.
Before adopting Gemini, CyberLink had used an open-source AI model to support this feature. But the model often fell short because it couldn’t correctly identify the locations or ambience in video clips. Lin notes that Gemini is able to understand video content well and assign correct attributes, producing edits that require little to no manual adjustment.
“The comprehension capability of Gemini is really impressive. With it powering auto-editing, our users can generate finished videos in seconds, with most of the heavy lifting already done,” he adds.
CyberLink also turned to Gemini to improve its image style change feature in its photo editing app. Earlier solutions couldn’t preserve the key features of original photos, often resulting in unrealistic results.
Following the technical advice of the Google Cloud team, CyberLink now employs Gemini 2.5 Flash to analyze the photos and users intent, then generate a tailored prompt. That prompt is sent to Imagen, which delivers the final output The team reports that more than 85% of the resulting images meet their quality standards.
“What’s amazing about Google Cloud is that we can easily combine different AI models to realize solutions tailored to our needs,” says Lin. “By merging the comprehension capability of Gemini with the image generation power of Imagen, we’re able to deliver a photo editing tool that can interact with users more naturally and realize creative ideas efficiently.”
By merging the comprehension capability of Gemini with the image generation power of Imagen, we’re able to deliver a photo editing tool that can interact with users more naturally and realize creative ideas efficiently.
Vincent Lin
President, CyberLink
Scaling up AI use for product and operation optimization
Next, aiming to become a comprehensive AI solution expert, CyberLink is expanding its use of generative AI internally. For example, Gemini is now part of its quality assurance workflow, helping to assess layout aesthetics and user experience across its software products.
While continuing to optimize its video auto-edit and image style change features, the team also plans to use Gemini to enhance document translation and streamline customer support.
“As AI continues to evolve, there are still plenty of implementation ideas for us to explore in the multimedia sector,” notes Lin. “Offering wide-ranging, top-notch AI solutions and global resources, Google Cloud is our great collaborator in this transformation journey.”
As AI continues to evolve, there are still plenty of implementation ideas for us to explore in the multimedia sector. Offering wide-ranging, top-notch AI solutions and global resources, Google Cloud is our great collaborator in this transformation journey.
Vincent Lin
President, CyberLink