Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Liu, Tao, Wan, Gang, Ren, Kan, Wen, Shibo
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2602.23141
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915820184010752
author Liu, Tao
Wan, Gang
Ren, Kan
Wen, Shibo
author_facet Liu, Tao
Wan, Gang
Ren, Kan
Wen, Shibo
contents We propose a new unsupervised framework for online video stabilization. Unlike methods based on deep learning that require paired stable and unstable datasets, our approach instantiates the classical stabilization pipeline with three stages and incorporates a multithreaded buffering mechanism. This design addresses three longstanding challenges in end-to-end learning: limited data, poor controllability, and inefficiency on hardware with constrained resources. Existing benchmarks focus mainly on handheld videos with a forward view in visible light, which restricts the applicability of stabilization to domains such as UAV nighttime remote sensing. To fill this gap, we introduce a new multimodal UAV aerial video dataset (UAV-Test). Experiments show that our method consistently outperforms state-of-the-art online stabilizers in both quantitative metrics and visual quality, while achieving performance comparable to offline methods.
format Preprint
id arxiv_https___arxiv_org_abs_2602_23141
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle No Labels, No Look-Ahead: Unsupervised Online Video Stabilization with Classical Priors
Liu, Tao
Wan, Gang
Ren, Kan
Wen, Shibo
Computer Vision and Pattern Recognition
We propose a new unsupervised framework for online video stabilization. Unlike methods based on deep learning that require paired stable and unstable datasets, our approach instantiates the classical stabilization pipeline with three stages and incorporates a multithreaded buffering mechanism. This design addresses three longstanding challenges in end-to-end learning: limited data, poor controllability, and inefficiency on hardware with constrained resources. Existing benchmarks focus mainly on handheld videos with a forward view in visible light, which restricts the applicability of stabilization to domains such as UAV nighttime remote sensing. To fill this gap, we introduce a new multimodal UAV aerial video dataset (UAV-Test). Experiments show that our method consistently outperforms state-of-the-art online stabilizers in both quantitative metrics and visual quality, while achieving performance comparable to offline methods.
title No Labels, No Look-Ahead: Unsupervised Online Video Stabilization with Classical Priors
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2602.23141