Agentic Very Long Video Understanding
This addresses the need for contextual understanding in always-on wearable devices, though it appears incremental by building on existing agentic and graph-based approaches.
The paper tackles the problem of understanding very long, continuous egocentric video streams for personal AI assistants by introducing EGAgent, an agentic framework using entity scene graphs, achieving state-of-the-art performance of 57.5% on EgoLifeQA and competitive 74.1% on Video-MME (Long).
The advent of always-on personal AI assistants, enabled by all-day wearable devices such as smart glasses, demands a new level of contextual understanding, one that goes beyond short, isolated events to encompass the continuous, longitudinal stream of egocentric video. Achieving this vision requires advances in long-horizon video understanding, where systems must interpret and recall visual and audio information spanning days or even weeks. Existing methods, including large language models and retrieval-augmented generation, are constrained by limited context windows and lack the ability to perform compositional, multi-hop reasoning over very long video streams. In this work, we address these challenges through EGAgent, an enhanced agentic framework centered on entity scene graphs, which represent people, places, objects, and their relationships over time. Our system equips a planning agent with tools for structured search and reasoning over these graphs, as well as hybrid visual and audio search capabilities, enabling detailed, cross-modal, and temporally coherent reasoning. Experiments on the EgoLifeQA and Video-MME (Long) datasets show that our method achieves state-of-the-art performance on EgoLifeQA (57.5%) and competitive performance on Video-MME (Long) (74.1%) for complex longitudinal video understanding tasks.