OACap：面向详细视频描述的字幕引导主动观察方法

View PDF HTML (experimental)

Abstract:Unlike question answering, where a query specifies a perceptual target, open-ended detailed captioning must determine what evidence to acquire and how to integrate it into a coherent global description. We propose Observation-Adaptive Video Captioner (OACap), a caption-centered agent that turns unsupported or missing information into targeted observation requests, uses spatiotemporal tools to acquire evidence, and iteratively verifies and revises the caption. OACap captions consistently improve performance across general video-understanding benchmarks using a fixed text-only reader. To better diagnose the information they preserve, we introduce Factual-Semantic Captioning Bench (FSCap-Bench). Its Core-Support Factuality Score (CSFS) combines claim precision with importance-aware fact coverage and responds as expected to four controlled perturbation types on 93.0-98.3% of valid caption pairs. Its Cross-Modal Semantic Round-Trip Score (CSRS) measures semantic preservation through a video-caption-video-caption round trip by matching atomic facts from source and recovered captions. Together, they connect downstream textual-memory utility with factual and semantic preservation.

Comments:	14 pages, 4 figures, 4 tables. Technical report
Subjects:	Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL); Multiagent Systems (cs.MA); Multimedia (cs.MM)
ACM classes:	I.2.10; I.2.7
Cite as:	arXiv:2610.29327 [cs.CV]
	(or arXiv:2610.29327v1 [cs.CV] for this version)
	https://doi.org/10.48550/arXiv.2610.29327 arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Hao Shi [view email]
[v1] Wed, 21 Oct 2026 07:26:26 UTC (2,948 KB)