Video-LLaVA: Learning United Visual Representation by Alignment Before Projection

TL;DR

Video-LLaVA introduces a unified visual representation approach for large vision-language models, enabling better multi-modal understanding and outperforming existing models on various image and video benchmarks.

Contribution

The paper proposes a unified visual representation for images and videos in LVLMs, improving multi-modal learning and performance across multiple benchmarks.

Findings

Outperforms Video-ChatGPT on several video benchmarks

Achieves superior results on 9 image question-answering datasets

Mutually benefits image and video understanding within a unified model

Abstract

The Large Vision-Language Model (LVLM) has enhanced the performance of various downstream tasks in visual-language understanding. Most existing approaches encode images and videos into separate feature spaces, which are then fed as inputs to large language models. However, due to the lack of unified tokenization for images and videos, namely misalignment before projection, it becomes challenging for a Large Language Model (LLM) to learn multi-modal interactions from several poor projection layers. In this work, we unify visual representation into the language feature space to advance the foundational LLM towards a unified LVLM. As a result, we establish a simple but robust LVLM baseline, Video-LLaVA, which learns from a mixed dataset of images and videos, mutually enhancing each other. Video-LLaVA achieves superior performances on a broad range of 9 image benchmarks across 5 image question-answering datasets and 4 image benchmark toolkits. Additionally, our Video-LLaVA also outperforms Video-ChatGPT by 5.8%, 9.9%, 18.6%, and 10.1% on MSRVTT, MSVD, TGIF, and ActivityNet, respectively. Notably, extensive experiments demonstrate that Video-LLaVA mutually benefits images and videos within a unified visual representation, outperforming models designed specifically for images or videos. We aim for this work to provide modest insights into the multi-modal inputs for the LLM. Code address: \href{https://github.com/PKU-YuanGroup/Video-LLaVA}