April 23, at the 2024 SenseTime Technology Exchange Day event held today, the Vincent video generation platform originally created by SenseTime made its debut, focusing on providing users with a controllable Vincent video creation experience. At the scene, Xu Li, chairman and CEO of SenseTime, showed three video clips completely generated by large models, which fully demonstrated the high degree of controllability of characters, actions and scenes by SenseTime's Vincent video technology.
The first paragraph uses a realistic style to show close-ups of the faces of different characters. The emotions and expressions of the characters in the picture change in every detail. The second segment is a "Blade Runner"-style sci-fi animation clip, which vividly depicts the subtle action interactions between different characters. The third segment is a 3D animation that incorporates elements of Chinese classical aesthetics. From the deep night with lanterns, to a leisurely stroll in the rain, to a girl dancing among flowers, it demonstrates the relationship between people, objects, and scenes that Shang Tang Wensheng video technology has deep understanding.
SenseTime’s Wensheng video generation platform comes from SenseTime’s long-term accumulation in the fields of generative AI technologies such as 3D generation, nerf, and digital humans, enabling precise and controllable control of characters, actions, and scenes. Through simple text input or description, the platform can quickly generate coherent and consistent video content, and the character's shape, scene, etc. can be customized. The
Shangtang Wensheng video platform will be officially released soon, which will provide efficient and convenient video generation services to customers in various industries and promote the widespread application and popularization of video generation technology.