OmAgent: A Multi-modal Agent Framework for Complex Video Understanding with Task Divide-and-Conquer

TL;DR

OmAgent introduces a multi-modal framework that efficiently processes and understands complex videos by combining intelligent frame retrieval, divide-and-conquer reasoning, and autonomous tool invocation, significantly improving video comprehension accuracy.

Contribution

It presents OmAgent, a novel system that enhances video understanding through efficient data management, autonomous reasoning, and dynamic API integration, addressing limitations of traditional methods.

Findings

OmAgent effectively handles 24-hour CCTV footage and full-length films.

It significantly reduces information loss compared to traditional frame extraction methods.

Experimental results demonstrate improved accuracy and robustness in complex video tasks.

Abstract

Recent advancements in Large Language Models (LLMs) have expanded their capabilities to multimodal contexts, including comprehensive video understanding. However, processing extensive videos such as 24-hour CCTV footage or full-length films presents significant challenges due to the vast data and processing demands. Traditional methods, like extracting key frames or converting frames to text, often result in substantial information loss. To address these shortcomings, we develop OmAgent, efficiently stores and retrieves relevant video frames for specific queries, preserving the detailed content of videos. Additionally, it features an Divide-and-Conquer Loop capable of autonomous reasoning, dynamically invoking APIs and tools to enhance query processing and accuracy. This approach ensures robust video understanding, significantly reducing information loss. Experimental results affirm OmAgent's efficacy in handling various types of videos and complex tasks. Moreover, we have endowed it with greater autonomy and a robust tool-calling system, enabling it to accomplish even more intricate tasks.