VideoRefer Suite: Advancing Spatial-Temporal Object Understanding with Video LLM

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Yuan, Yuqian, Zhang, Hang, Li, Wentong, Cheng, Zesen, Zhang, Boqiang, Li, Long, Li, Xin, Zhao, Deli, Zhang, Wenqiao, Zhuang, Yueting, Zhu, Jianke, Bing, Lidong
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908282506969088
author Yuan, Yuqian
Zhang, Hang
Li, Wentong
Cheng, Zesen
Zhang, Boqiang
Li, Long
Li, Xin
Zhao, Deli
Zhang, Wenqiao
Zhuang, Yueting
Zhu, Jianke
Bing, Lidong
author_facet Yuan, Yuqian
Zhang, Hang
Li, Wentong
Cheng, Zesen
Zhang, Boqiang
Li, Long
Li, Xin
Zhao, Deli
Zhang, Wenqiao
Zhuang, Yueting
Zhu, Jianke
Bing, Lidong
contents Video Large Language Models (Video LLMs) have recently exhibited remarkable capabilities in general video understanding. However, they mainly focus on holistic comprehension and struggle with capturing fine-grained spatial and temporal details. Besides, the lack of high-quality object-level video instruction data and a comprehensive benchmark further hinders their advancements. To tackle these challenges, we introduce the VideoRefer Suite to empower Video LLM for finer-level spatial-temporal video understanding, i.e., enabling perception and reasoning on any objects throughout the video. Specially, we thoroughly develop VideoRefer Suite across three essential aspects: dataset, model, and benchmark. Firstly, we introduce a multi-agent data engine to meticulously curate a large-scale, high-quality object-level video instruction dataset, termed VideoRefer-700K. Next, we present the VideoRefer model, which equips a versatile spatial-temporal object encoder to capture precise regional and sequential representations. Finally, we meticulously create a VideoRefer-Bench to comprehensively assess the spatial-temporal understanding capability of a Video LLM, evaluating it across various aspects. Extensive experiments and analyses demonstrate that our VideoRefer model not only achieves promising performance on video referring benchmarks but also facilitates general video understanding capabilities.
format Preprint
id arxiv_https___arxiv_org_abs_2501_00599
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle VideoRefer Suite: Advancing Spatial-Temporal Object Understanding with Video LLM
Yuan, Yuqian
Zhang, Hang
Li, Wentong
Cheng, Zesen
Zhang, Boqiang
Li, Long
Li, Xin
Zhao, Deli
Zhang, Wenqiao
Zhuang, Yueting
Zhu, Jianke
Bing, Lidong
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Video Large Language Models (Video LLMs) have recently exhibited remarkable capabilities in general video understanding. However, they mainly focus on holistic comprehension and struggle with capturing fine-grained spatial and temporal details. Besides, the lack of high-quality object-level video instruction data and a comprehensive benchmark further hinders their advancements. To tackle these challenges, we introduce the VideoRefer Suite to empower Video LLM for finer-level spatial-temporal video understanding, i.e., enabling perception and reasoning on any objects throughout the video. Specially, we thoroughly develop VideoRefer Suite across three essential aspects: dataset, model, and benchmark. Firstly, we introduce a multi-agent data engine to meticulously curate a large-scale, high-quality object-level video instruction dataset, termed VideoRefer-700K. Next, we present the VideoRefer model, which equips a versatile spatial-temporal object encoder to capture precise regional and sequential representations. Finally, we meticulously create a VideoRefer-Bench to comprehensively assess the spatial-temporal understanding capability of a Video LLM, evaluating it across various aspects. Extensive experiments and analyses demonstrate that our VideoRefer model not only achieves promising performance on video referring benchmarks but also facilitates general video understanding capabilities.
title VideoRefer Suite: Advancing Spatial-Temporal Object Understanding with Video LLM
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2501.00599