AlanaVLM: A Multimodal Embodied AI Foundation Model for Egocentric Video Understanding

TL;DR

AlanaVLM is a 7-billion-parameter multimodal model trained on egocentric videos, achieving state-of-the-art results in embodied video question answering and advancing AI's ability to understand human-centric environments.

Contribution

The paper introduces EVUD dataset, proposes AlanaVLM model, and demonstrates its superior performance on embodied video understanding benchmarks.

Findings

Achieves 3.6% better performance than GPT-4-based models on OpenEQA.

Outperforms Claude 3 and Gemini Pro Vision 1.0 in embodied video question answering.

Shows competitive results with GPT-4V, especially in spatial reasoning.

Abstract

AI personal assistants deployed via robots or wearables require embodied understanding to collaborate with humans effectively. However, current Vision-Language Models (VLMs) primarily focus on third-person view videos, neglecting the richness of egocentric perceptual experience. To address this gap, we propose three key contributions. First, we introduce the Egocentric Video Understanding Dataset (EVUD) for training VLMs on video captioning and question answering tasks specific to egocentric videos. Second, we present AlanaVLM, a 7B parameter VLM trained using parameter-efficient methods on EVUD. Finally, we evaluate AlanaVLM's capabilities on OpenEQA, a challenging benchmark for embodied video question answering. Our model achieves state-of-the-art performance, outperforming open-source models including strong Socratic models using GPT-4 as a planner by 3.6%. Additionally, we outperform Claude 3 and Gemini Pro Vision 1.0 and showcase competitive results compared to Gemini Pro 1.5 and GPT-4V, even surpassing the latter in spatial reasoning. This research paves the way for building efficient VLMs that can be deployed in robots or wearables, leveraging embodied video understanding to collaborate seamlessly with humans in everyday tasks, contributing to the next generation of Embodied AI.