腾讯日前发布了视频生成模型VideoCrafter2的升级版,在光影效果等方面有了大幅度的提高。新版模型可以根据用户的文字描述生成高质量的几秒钟视频,相比之前的版本在画面质量、人物动作等方面改进明显,生成视频的逼真度也有了很大提升。
业内人士表示,这种视频AI生成技术的进步,将为短视频、动画等行业带来更多应用场景。同时也提醒相关从业者要加强内容审核,防止深度假信息的生成和传播。
Title: Tencent Upgrades Video AI Generation Model for Higher Realism
Keywords: video generation, AI model, realism
News content: Tencent has released an upgraded version of its video generation model VideoCrafter2, with significant improvements in lighting, shadows and other aspects. The new model can generate high-quality videos of a few seconds based on users’ text descriptions. Compared to the previous version, there are noticeable enhancements in image quality, character motions, etc., making the generated videos much more realistic.
Industry insiders said the advancement of such video AI generation technology will bring more application scenarios for short videos, animations and other industries. At the same time, it also reminds related practitioners to strengthen content review to prevent the generation and dissemination of deepfakes.
【来源】https://www.pingwest.com/w/291981
Views: 1