Unifying Specialized Visual Encoders for Video Language Models

TL;DR

MERV enhances video understanding in VideoLLMs by integrating multiple specialized frozen visual encoders, leading to improved accuracy, faster training, and richer visual representations compared to single-encoder approaches.

Contribution

This paper introduces MERV, a novel multi-encoder framework that unifies diverse visual features for VideoLLMs, surpassing prior methods in accuracy and efficiency.

Findings

Up to 3.7% accuracy improvement over Video-LLaVA

2.2% better zero-shot Perception Test accuracy than SeViLA

Faster training with minimal parameter increase

Abstract

The recent advent of Large Language Models (LLMs) has ushered sophisticated reasoning capabilities into the realm of video through Video Large Language Models (VideoLLMs). However, VideoLLMs currently rely on a single vision encoder for all of their visual processing, which limits the amount and type of visual information that can be conveyed to the LLM. Our method, MERV, Multi-Encoder Representation of Videos, instead leverages multiple frozen visual encoders to create a unified representation of a video, providing the VideoLLM with a comprehensive set of specialized visual knowledge. Spatio-temporally aligning the features from each encoder allows us to tackle a wider range of open-ended and multiple-choice video understanding questions and outperform prior state-of-the-art works. MERV is up to 3.7% better in accuracy than Video-LLaVA across the standard suite video understanding benchmarks, while also having a better Video-ChatGPT score. We also improve upon SeViLA, the previous best on zero-shot Perception Test accuracy, by 2.2%. MERV introduces minimal extra parameters and trains faster than equivalent single-encoder methods while parallelizing the visual processing. Finally, we provide qualitative evidence that MERV successfully captures domain knowledge from each of its encoders. Our results offer promising directions in utilizing multiple vision encoders for comprehensive video understanding.