在中关村论坛未来人工智能先锋论坛上，生数科技与清华大学联合宣布了一项重大突破——正式发布了中国首个长时长、高一致性、高动态性视频大模型「Vidu」。
该模型的发布标志着中国在人工智能领域的又一重要进展。此次推出的「Vidu」模型，采用了团队独创的Diffusion与Transformer融合的架构U-ViT，不仅能够生成高质量的视频内容，还支持一键生成长达16秒、分辨率高达1080P的高清视频。这一技术的推出，将为影视制作、广告宣传、教育培训等领域带来革命性的改变。
据了解，「Vidu」大模型的研发历时多年，汇聚了众多人工智能和计算机视觉领域的顶尖专家。其独特的架构保证了生成视频的连贯性和动态性，为用户提供了更加逼真的视觉体验。同时，「Vidu」模型的推出也将加速中国在全球人工智能领域的竞争力和影响力。
生数科技表示，未来他们将继续深耕人工智能领域，与各方携手推动人工智能技术的普及和应用，为社会发展贡献力量。此次「Vidu」模型的发布，无疑将引发全球范围内的关注和讨论。
以上内容转载自生数科技。
英语如下：
News Title: “Shengshu Technology and Tsinghua University Jointly Create China’s First Long Video Large Model ‘Vidu'”
Keywords: Shengshu Technology, Tsinghua University, Vidu, Long Video Large Model
News Content:
Shengshu Technology and Tsinghua University have jointly announced the release of China’s first long-duration, high-consistency, and high-dynamic video large model “Vidu” at the Zhongguancun Forum Future AI Vanguard Forum.
The release of this model marks another significant advancement in China’s artificial intelligence sector. The new “Vidu” model, utilizing the team’s unique architecture U-ViT that combines Diffusion and Transformer, is not only capable of generating high-quality video content but also supports one-click generation of videos up to 16 seconds long with a resolution of up to 1080P. The introduction of this technology will bring revolutionary changes to the fields of film production, advertising, education, and training.
It is understood that the development of the “Vidu” large model has taken years and has gathered many top experts in artificial intelligence and computer vision. Its unique architecture ensures the continuity and dynamic nature of the generated videos, providing users with a more realistic visual experience. At the same time, the launch of the “Vidu” model will also accelerate China’s competitiveness and influence in the global artificial intelligence field.
Shengshu Technology expressed that in the future, they will continue to focus on the field of artificial intelligence, work with all parties to promote the popularization and application of artificial intelligence technology, and contribute to social development. The release of the “Vidu” model will undoubtedly attract global attention and discussion.
The content above is translated from Shengshu Technology.
【来源】https://ai-bot.cn/go/?url=aHR0cHM6Ly9tcC53ZWl4aW4ucXEuY29tL3MvTGJhX1dKVFFXalEzbE9hTE5KQTRxQQ%3D%3D
Views: 1