ShengShu Technology, in collaboration with Tsinghua University, has launched Vidu, a cutting-edge large video generation model, making waves in the global AI landscape. Designed to cater to diverse creative needs, Vidu offers both text-to-video and image-to-video generation capabilities, empowering users to produce high-quality video content with ease.
One of Vidu's standout features is its impressive speed, enabling the creation of 4-second clips in just 30 seconds and generating 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 and surrealistic content with depth and complexity,\" explained Zhu Jun, deputy director of the Tsinghua Institute for Artificial Intelligence.
Vidu excels in various genres, including sci-fi, romance, and animation, capturing the essence of each style while incorporating high-quality cinematic effects like smoke and lens flares. The AI model is adept at managing different shot types, such as long shots, close-ups, and medium shots, and can effortlessly produce effects like long takes, focus pulls, and smooth scene transitions.
Enhancing creative freedom, Vidu allows users to upload portraits or customized character images and use text descriptions to direct characters to perform any action in any scene. This streamlined video production process is a game-changer for content creators, entrepreneurs, and young professionals looking to innovate in the digital space.
The core architecture of Vidu was proposed in early 2022 and was unveiled at the 2024 Zhongguancun Forum in Beijing in April, shortly after OpenAI announced its Sora video model. Since its debut, Vidu has maintained a low profile, while similar tools like Kuaishou's generative video model Kling and the ChatGLM large language model family have been increasingly adopted by users.
As Vidu becomes available for global use, it positions itself as a formidable player in the AI-driven video generation market, offering dynamic and versatile solutions for a wide range of applications.
Reference(s):
China-developed large video generation model available for global use
cgtn.com
