paper-with-me

ActivityNet-QA

홈페이지 · 논문 146편

The ActivityNet-QA dataset contains 58,000 human-annotated QA pairs on 5,800 videos derived from the popular ActivityNet dataset. The dataset provides a benchmark for testing the performance of VideoQA models on long-term spatio-temporal reasoning. Source: [ActivityNet-QA: A Dataset for Understanding Complex Web Videos via Question Answering](/paper/activitynet-qa-a-dataset-for-understanding)

VideosTexts English

벤치마크

Zero-Shot Video Question Answer on ActivityNet-QA 결과 56개
Video Question Answering on ActivityNet-QA 결과 36개