China’s cutting-edge artificial intelligence company ShengShu Technology, in collaboration with Tsinghua University, has launched “Vidu”, a large-scale video generation model now available for global use. This groundbreaking AI model offers text-to-video and image-to-video generation capabilities, marking a significant leap in digital content creation.
Vidu stands out with its remarkable efficiency, capable of producing 4-second video clips in just 30 seconds, and can generate videos up to 32 seconds long in a single instance. “Vidu can simulate the real physical world, creating detailed scenes that adhere to physical laws, such as natural lighting and shadow effects, as well as intricate facial expressions. Additionally, it can generate surrealistic content with depth and complexity,” said Zhu Jun, deputy director of the Tsinghua Institute for Artificial Intelligence.
The model excels across various genres including science fiction, romance, and animation, producing scenes that capture the essence of each style. It also delivers high-quality cinematic effects such as smoke and lens flares, enhancing the visual experience. Vidu adeptly manages different shot types, from long shots to close-ups, and effortlessly produces effects like long takes, focus pulls, and smooth scene transitions.
A standout feature of Vidu is its ability to allow users to upload portraits or customized character images. By using text descriptions, users can direct these characters to perform any action in any scene, streamlining the video production process and expanding creative freedom.
According to ShengShu Technology, Vidu’s core architecture was proposed in 2022. The AI model was unveiled at the 2024 Zhongguancun Forum in Beijing in April, two months after OpenAI announced its Sora video model. Despite its early development, Vidu has kept a low profile since the forum.
In the interim, similar tools such as Kuaishou’s generative video model Kling and the large language model family ChatGLM have become available to users, highlighting the rapid advancement in AI-driven video technology.
Reference(s):
China-developed large video generation model available for global use
cgtn.com
