We present HourVideo, a benchmark dataset for hour-long video-language understanding. Our dataset consists of a novel task suite comprising summarization, perception (recall, tracking), visual reasoning (spatial, temporal, predictive, causal, counterfactual), and navigation (room-to-room, object retrieval) tasks. HourVideo includes 500 manually curated egocentric videos from the Ego4D dataset, spanning durations of 20 to 120 minutes, and features 12,976 high-quality, five-way multiple-choice questions. Benchmarking results reveal that multimodal models, including GPT-4 and LLaVA-NeXT, achieve marginal improvements over random chance. In stark contrast, human experts significantly outperform the state-of-the-art long-context multimodal model, Gemini Pro 1.5 (85.0% vs. 37.3%), highlighting a substantial gap in multimodal capabilities. Our benchmark, evaluation toolkit, prompts, and documentation are available at https://hourvideo.stanford.edu
@article{arxiv.2411.04998,
title = {HourVideo: 1-Hour Video-Language Understanding},
author = {Keshigeyan Chandrasegaran and Agrim Gupta and Lea M. Hadzic and Taran Kota and Jimming He and Cristóbal Eyzaguirre and Zane Durante and Manling Li and Jiajun Wu and Li Fei-Fei},
journal= {arXiv preprint arXiv:2411.04998},
year = {2024}
}
Comments
NeurIPS 2024 Datasets and Benchmarks Track; 28 pages