The Chinese internet giant Tencent introduced an artificial intelligence (AI) model on March 15, called Follow-Your-Click.
The new animation tool, Follow-Your-Click, has appeared on Microsoft’s open-source GitHub page. This tool enables users to create short animated videos by clicking on parts of an image and transforming them into dynamic videos.
Some images and videos created by Follow-Your-Click. (Photo: AIbase)
This project is the result of a collaboration between Tencent’s Hunyuan team, the Hong Kong University of Science and Technology, and Tsinghua University in China.
Tencent plans to release the full source code for this tool in April, but a trial version is already available on GitHub.
Follow-Your-Click was developed to address issues faced by other models that convert images into videos, which tend to move the entire scene instead of focusing on specific objects within the images.
Other AI models require users to provide detailed descriptions of how and where they want the images to move.
Researchers wrote in a post on arXiv: “Our framework enables simpler yet more precise user control and better generation performance compared to previous methods.”
Creating videos from text has become a hot topic since OpenAI introduced Sora on February 15.
Earlier, in January, Tencent launched an open-source video editing and creation tool called VideoCrafter2, capable of generating videos from text. This is an updated version of VideoCrafter1, released in October 2023, but was limited to 2-second videos.
Tencent’s competitors in China have also joined the race. Alibaba recently launched a portrait video creation tool named EMO, which transforms images and audio prompts into singing and speaking videos.
Around the same time, ByteDance released a text-to-video model called MagicVideo-V2.
