Skip to yearly menu bar Skip to main content

Spotlight Poster

InternVid: A Large-scale Video-Text Dataset for Multimodal Understanding and Generation

Yi Wang · Yinan He · Yizhuo Li · Kunchang Li · Jiashuo Yu · Xin Ma · Xinhao Li · Guo Chen · Xinyuan Chen · Yaohui Wang · Ping Luo · Ziwei Liu · Yali Wang · Limin Wang · Yu Qiao

Halle B #198
[ ] [ Project Page ]
Thu 9 May 1:45 a.m. PDT — 3:45 a.m. PDT


This paper introduces InternVid, a large-scale video-centric multimodal dataset that enables learning powerful and transferable video-text representations for multimodal understanding and generation. InternVid contains over 7 million videos lasting nearly 760K hours, yielding 234M video clips accompanied by detailed descriptions of total 4.1B words. Our core contribution is to develop a scalable approach to autonomously build a high-quality video-text dataset with large language models (LLM), thereby showcasing its efficacy in learning video-language representation at scale. Specifically, we utilize a multi-scale approach to generate video-related descriptions. Furthermore, we introduce ViCLIP, a video-text representation learning model based on ViT-L. Learned on InternVid via contrastive learning, this model demonstrates leading zero-shot action recognition and competitive video retrieval performance. Beyond basic video understanding tasks like recognition and retrieval, our dataset and model have broad applications. They are particularly beneficial for generating interleaved video-text data for learning a video-centric dialogue system, advancing video-to-text and text-to-video generation research. These proposed resources provide a tool for researchers and practitioners interested in multimodal video understanding and generation.

Chat is not available.