HourVideo

홈페이지 · 논문 5편

We introduce HourVideo, a benchmark dataset for hour-long video-language understanding. HourVideo 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. We hope to establish HourVideo as a benchmark challenge to spur the development of advanced multimodal models capable of truly understanding endless streams of visual data.