Saved in:
Bibliographic Details
Main Authors: Zou, Heqing, Luo, Tianze, Xie, Guiyang, Zhang, Victor Xiao Jie, Lv, Fengmao, Wang, Guangcong, Chen, Junyang, Wang, Zhuochen, Zhang, Hansheng, Zhang, Huaijian
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2501.01645
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916734123900928
author Zou, Heqing
Luo, Tianze
Xie, Guiyang
Zhang, Victor Xiao Jie
Lv, Fengmao
Wang, Guangcong
Chen, Junyang
Wang, Zhuochen
Zhang, Hansheng
Zhang, Huaijian
author_facet Zou, Heqing
Luo, Tianze
Xie, Guiyang
Zhang, Victor Xiao Jie
Lv, Fengmao
Wang, Guangcong
Chen, Junyang
Wang, Zhuochen
Zhang, Hansheng
Zhang, Huaijian
contents Multimodal large language models have become a popular topic in deep visual understanding due to many promising real-world applications. However, hour-long video understanding, spanning over one hour and containing tens of thousands of visual frames, remains under-explored because of 1) challenging long-term video analyses, 2) inefficient large-model approaches, and 3) lack of large-scale benchmark datasets. Among them, in this paper, we focus on building a large-scale hour-long long video benchmark, HLV-1K, designed to evaluate long video understanding models. HLV-1K comprises 1009 hour-long videos with 14,847 high-quality question answering (QA) and multi-choice question asnwering (MCQA) pairs with time-aware query and diverse annotations, covering frame-level, within-event-level, cross-event-level, and long-term reasoning tasks. We evaluate our benchmark using existing state-of-the-art methods and demonstrate its value for testing deep long video understanding capabilities at different levels and for various tasks. This includes promoting future long video understanding tasks at a granular level, such as deep understanding of long live videos, meeting recordings, and movies.
format Preprint
id arxiv_https___arxiv_org_abs_2501_01645
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle HLV-1K: A Large-scale Hour-Long Video Benchmark for Time-Specific Long Video Understanding
Zou, Heqing
Luo, Tianze
Xie, Guiyang
Zhang, Victor Xiao Jie
Lv, Fengmao
Wang, Guangcong
Chen, Junyang
Wang, Zhuochen
Zhang, Hansheng
Zhang, Huaijian
Computer Vision and Pattern Recognition
Artificial Intelligence
Multimodal large language models have become a popular topic in deep visual understanding due to many promising real-world applications. However, hour-long video understanding, spanning over one hour and containing tens of thousands of visual frames, remains under-explored because of 1) challenging long-term video analyses, 2) inefficient large-model approaches, and 3) lack of large-scale benchmark datasets. Among them, in this paper, we focus on building a large-scale hour-long long video benchmark, HLV-1K, designed to evaluate long video understanding models. HLV-1K comprises 1009 hour-long videos with 14,847 high-quality question answering (QA) and multi-choice question asnwering (MCQA) pairs with time-aware query and diverse annotations, covering frame-level, within-event-level, cross-event-level, and long-term reasoning tasks. We evaluate our benchmark using existing state-of-the-art methods and demonstrate its value for testing deep long video understanding capabilities at different levels and for various tasks. This includes promoting future long video understanding tasks at a granular level, such as deep understanding of long live videos, meeting recordings, and movies.
title HLV-1K: A Large-scale Hour-Long Video Benchmark for Time-Specific Long Video Understanding
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2501.01645