KeyVideoLLM: Towards Large-scale Video Keyframe Selection

TL;DR

KeyVideoLLM is a novel keyframe selection method that significantly reduces data size, improves processing speed, and enhances video understanding tasks, achieving state-of-the-art results in large-scale video datasets.

Contribution

It introduces a text-video frame similarity-based approach for efficient, robust, and hyperparameter-free keyframe selection tailored for VideoLLMs.

Findings

Achieves up to 60.9x data compression

Enhances processing speed by up to 200x

Attains state-of-the-art results in video question-answering

Abstract

Recently, with the rise of web videos, managing and understanding large-scale video datasets has become increasingly important. Video Large Language Models (VideoLLMs) have emerged in recent years due to their strong video understanding capabilities. However, training and inference processes for VideoLLMs demand vast amounts of data, presenting significant challenges to data management, particularly regarding efficiency, robustness, and effectiveness. In this work, we present KeyVideoLLM, a text-video frame similarity-based keyframe selection method designed to manage VideoLLM data efficiently, robustly, and effectively. Specifically, KeyVideoLLM achieves a remarkable data compression rate of up to 60.9 times, substantially lowering disk space requirements, which proves its high efficiency. Additionally, it maintains a 100% selection success rate across all video formats and scales, enhances processing speed by up to 200 times compared to existing keyframe selection methods, and does not require hyperparameter tuning. Beyond its outstanding efficiency and robustness, KeyVideoLLM further improves model performance in video question-answering tasks during both training and inference stages. Notably, it consistently achieved the state-of-the-art (SoTA) experimental results on diverse datasets.