在今日的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共同宣布了一项重大科技创新——中国首个长时长视频大模型「Vidu」的发布。这款模型标志着中国在人工智能视频生成领域取得的重大突破，为数字内容创作带来了全新的可能性。
「Vidu」模型基于生数科技团队的原创技术，巧妙地融合了Diffusion算法与Transformer架构，构建出先进的U-ViT框架。这一创新设计使得「Vidu」具备了高一致性与高动态性的特点，能够支持一键生成长达16秒、分辨率高达1080P的高清视频内容。这不仅极大地提升了视频生成的效率，也保证了生成视频的质量和连贯性。
这一合作项目彰显了中国在人工智能研究与应用上的领先地位，以及学术界与产业界紧密合作的成果。生数科技与清华大学的联合研发，无疑将推动中国乃至全球的视频内容创作进入一个全新的时代，预示着未来媒体、娱乐、教育等领域的视频生成和编辑将变得更加智能化和便捷。
随着「Vidu」的发布，我们期待看到更多创新应用的涌现，以及人工智能在视频内容创作领域带来的革命性变革。生数科技与清华大学的这一里程碑式创新，无疑将为我国人工智能产业的发展注入新的活力。
英语如下：
**News Title:** “SengiTech Teams Up with Tsinghua University to Launch China’s First Long-form Video Large Model ‘Vidu’: Revolutionizing Video Generation Technology”
**Keywords:** Vidu Launch, Tsinghua University Collaboration, Long-form Videos
**News Content:**
Title: SengiTech and Tsinghua University Join Forces to Unveil China’s First Long-form Video Large Model ‘Vidu’
At the Future Artificial Intelligence Pioneers Forum of the Zhongguancun Forum today, SengiTech and Tsinghua University jointly announced a major breakthrough in technological innovation – the release of China’s first long-form video large model, ‘Vidu’. This marks a significant advancement in China’s artificial intelligence (AI) video generation field, opening up new possibilities for digital content creation.
Built upon the original technology of the SengiTech team, ‘Vidu’ ingeniously combines Diffusion algorithms with Transformer architecture to create an advanced U-ViT framework. This innovative design endows ‘Vidu’ with high consistency and dynamism, enabling the generation of high-definition videos up to 16 seconds long and 1080P resolution with a single click. It not only significantly boosts video generation efficiency but also ensures the quality and continuity of the generated content.
This collaborative project demonstrates China’s leading position in AI research and application, as well as the fruits of close cooperation between academia and industry. The joint development by SengiTech and Tsinghua University is set to propel China and the global video content creation into a new era, foreshadowing a more intelligent and convenient future for video generation and editing across sectors such as media, entertainment, and education.
With the launch of ‘Vidu’, we anticipate the emergence of more innovative applications and a transformative shift in the video content creation landscape driven by AI. This milestone innovation from SengiTech and Tsinghua University will undoubtedly infuse new vitality into China’s AI industry.
【来源】https://mp.weixin.qq.com/s/Lba_WJTQWjQ3lOaLNJA4qA
Views: 1