InternVideo2: Scaling Foundation Models for Multimodal Video Understanding
This addresses the problem of comprehensive video understanding for AI applications, representing a significant but incremental advancement in scaling foundation models.
The paper tackles multimodal video understanding by introducing InternVideo2, a family of video foundation models that achieve state-of-the-art results in video recognition, video-text tasks, and video-centric dialogue, outperforming others on over 60 tasks and benchmarks for long video understanding.
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/.