VidEgoThink: Assessing Egocentric Video Understanding Capabilities for Embodied AI

TL;DR

VidEgoThink is a new benchmark designed to evaluate egocentric video understanding in Embodied AI, revealing current foundation models' limitations and highlighting the need for further advancements in this domain.

Contribution

The paper introduces VidEgoThink, a comprehensive benchmark with an automatic data generation pipeline for assessing egocentric video understanding capabilities in Embodied AI.

Findings

All tested MLLMs perform poorly on egocentric tasks.

The benchmark reveals significant gaps in current models' understanding.

Foundation models need substantial improvements for first-person scenarios.

Abstract

Recent advancements in Multi-modal Large Language Models (MLLMs) have opened new avenues for applications in Embodied AI. Building on previous work, EgoThink, we introduce VidEgoThink, a comprehensive benchmark for evaluating egocentric video understanding capabilities. To bridge the gap between MLLMs and low-level control in Embodied AI, we design four key interrelated tasks: video question-answering, hierarchy planning, visual grounding and reward modeling. To minimize manual annotation costs, we develop an automatic data generation pipeline based on the Ego4D dataset, leveraging the prior knowledge and multimodal capabilities of GPT-4o. Three human annotators then filter the generated data to ensure diversity and quality, resulting in the VidEgoThink benchmark. We conduct extensive experiments with three types of models: API-based MLLMs, open-source image-based MLLMs, and open-source video-based MLLMs. Experimental results indicate that all MLLMs, including GPT-4o, perform poorly across all tasks related to egocentric video understanding. These findings suggest that foundation models still require significant advancements to be effectively applied to first-person scenarios in Embodied AI. In conclusion, VidEgoThink reflects a research trend towards employing MLLMs for egocentric vision, akin to human capabilities, enabling active observation and interaction in the complex real-world environments.