Scholarlink

<p>Despite remarkable recent progress, existing long-form VideoQA datasets fall short of meeting the criteria for genuine long-form video understanding. This is primarily due to the use of short videos for question curation, and the reliance on limited-length sub-clips as clues to answer those questions. Meanwhile, previous datasets have limited focus on question type and modality. To remedy this, we introduce <b>LvBench</b>, a <b>L</b>ong-form <b>V</b>ideo understanding <b>Bench</b>mark for versatile multi-modal question-answering. Our LvBench stands out from existing long-form VideoQA datasets through three key characteristics: <Emphasis Type="BoldItalic">1) Extended temporal durations</Emphasis>: We consider videos ranging from 70 seconds to 4 hours, covering single-scene, multi-scene, and full-scene contexts. This design accounts for both video and clue lengths, capturing diverse contextual dynamics. <Emphasis Type="BoldItalic">2) Diverse question types and modalities</Emphasis>: LvBench introduces six distinct question types that evaluate various perceptual and cognitive capabilities, utilizing both video frames and subtitles. <Emphasis Type="BoldItalic">3) High-quality annotations</Emphasis>: We employ rigorous manual labeling by human annotators. Our dataset comprises 20,061 question-answer pairs sourced from 100 carefully selected movies across diverse genres, annotated collaboratively by multiple individuals. Analysis involving various baselines reveals a consistent trend: the performance of all existing methods significantly deteriorates when video and clue length increases. We expect LvBench to serve as a valuable resource for future works on long-form video understanding.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

LvBench: A Benchmark for Long-form Video Understanding with Versatile Multi-modal Question Answering

  • 期刊论文

  • 发表日期：

    2025年8月8日

查看全文
  • Hongjie Zhang,
  • Lu Dong,
  • Yi Liu,
  • Yifei Huang,
  • Yali Wang,
  • Limin Wang,
  • Yu Qiao

摘要

Despite remarkable recent progress, existing long-form VideoQA datasets fall short of meeting the criteria for genuine long-form video understanding. This is primarily due to the use of short videos for question curation, and the reliance on limited-length sub-clips as clues to answer those questions. Meanwhile, previous datasets have limited focus on question type and modality. To remedy this, we introduce LvBench, a Long-form Video understanding Benchmark for versatile multi-modal question-answering. Our LvBench stands out from existing long-form VideoQA datasets through three key characteristics: 1) Extended temporal durations: We consider videos ranging from 70 seconds to 4 hours, covering single-scene, multi-scene, and full-scene contexts. This design accounts for both video and clue lengths, capturing diverse contextual dynamics. 2) Diverse question types and modalities: LvBench introduces six distinct question types that evaluate various perceptual and cognitive capabilities, utilizing both video frames and subtitles. 3) High-quality annotations: We employ rigorous manual labeling by human annotators. Our dataset comprises 20,061 question-answer pairs sourced from 100 carefully selected movies across diverse genres, annotated collaboratively by multiple individuals. Analysis involving various baselines reveals a consistent trend: the performance of all existing methods significantly deteriorates when video and clue length increases. We expect LvBench to serve as a valuable resource for future works on long-form video understanding.