InternVideo2: Scaling Foundation Models for Multimodal Video Understanding

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Wang, Yi, Li, Kunchang, Li, Xinhao, Yu, Jiashuo, He, Yinan, Wang, Chenting, Chen, Guo, Pei, Baoqi, Yan, Ziang, Zheng, Rongkun, Xu, Jilan, Wang, Zun, Shi, Yansong, Jiang, Tianxiang, Li, Songze, Zhang, Hongjie, Huang, Yifei, Qiao, Yu, Wang, Yali, Wang, Limin
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866916356946919424
author Wang, Yi
Li, Kunchang
Li, Xinhao
Yu, Jiashuo
He, Yinan
Wang, Chenting
Chen, Guo
Pei, Baoqi
Yan, Ziang
Zheng, Rongkun
Xu, Jilan
Wang, Zun
Shi, Yansong
Jiang, Tianxiang
Li, Songze
Zhang, Hongjie
Huang, Yifei
Qiao, Yu
Wang, Yali
Wang, Limin
author_facet Wang, Yi
Li, Kunchang
Li, Xinhao
Yu, Jiashuo
He, Yinan
Wang, Chenting
Chen, Guo
Pei, Baoqi
Yan, Ziang
Zheng, Rongkun
Xu, Jilan
Wang, Zun
Shi, Yansong
Jiang, Tianxiang
Li, Songze
Zhang, Hongjie
Huang, Yifei
Qiao, Yu
Wang, Yali
Wang, Limin
contents We introduce InternVideo2, a new family of video foundation models (ViFM) that achieve the state-of-the-art results in video recognition, video-text tasks, and video-centric dialogue. Our core design is a progressive training approach that unifies the masked video modeling, crossmodal contrastive learning, and next token prediction, scaling up the video encoder size to 6B parameters. At the data level, we prioritize spatiotemporal consistency by semantically segmenting videos and generating video-audio-speech captions. This improves the alignment between video and text. Through extensive experiments, we validate our designs and demonstrate superior performance on over 60 video and audio tasks. Notably, our model outperforms others on various video-related dialogue and long video understanding benchmarks, highlighting its ability to reason and comprehend longer contexts. Code and models are available at https://github.com/OpenGVLab/InternVideo/tree/main/InternVideo2/.
format Preprint
id arxiv_https___arxiv_org_abs_2403_15377
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle InternVideo2: Scaling Foundation Models for Multimodal Video Understanding
Wang, Yi
Li, Kunchang
Li, Xinhao
Yu, Jiashuo
He, Yinan
Wang, Chenting
Chen, Guo
Pei, Baoqi
Yan, Ziang
Zheng, Rongkun
Xu, Jilan
Wang, Zun
Shi, Yansong
Jiang, Tianxiang
Li, Songze
Zhang, Hongjie
Huang, Yifei
Qiao, Yu
Wang, Yali
Wang, Limin
Computer Vision and Pattern Recognition
We introduce InternVideo2, a new family of video foundation models (ViFM) that achieve the state-of-the-art results in video recognition, video-text tasks, and video-centric dialogue. Our core design is a progressive training approach that unifies the masked video modeling, crossmodal contrastive learning, and next token prediction, scaling up the video encoder size to 6B parameters. At the data level, we prioritize spatiotemporal consistency by semantically segmenting videos and generating video-audio-speech captions. This improves the alignment between video and text. Through extensive experiments, we validate our designs and demonstrate superior performance on over 60 video and audio tasks. Notably, our model outperforms others on various video-related dialogue and long video understanding benchmarks, highlighting its ability to reason and comprehend longer contexts. Code and models are available at https://github.com/OpenGVLab/InternVideo/tree/main/InternVideo2/.
title InternVideo2: Scaling Foundation Models for Multimodal Video Understanding
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2403.15377