SOTA: Spike-Navigated Optimal TrAnsport Saliency Region Detection in Composite-bias Videos

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Liu, Wenxuan, Deng, Yao, Chen, Kang, Zhong, Xian, Yu, Zhaofei, Huang, Tiejun
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915268918247424
author Liu, Wenxuan
Deng, Yao
Chen, Kang
Zhong, Xian
Yu, Zhaofei
Huang, Tiejun
author_facet Liu, Wenxuan
Deng, Yao
Chen, Kang
Zhong, Xian
Yu, Zhaofei
Huang, Tiejun
contents Existing saliency detection methods struggle in real-world scenarios due to motion blur and occlusions. In contrast, spike cameras, with their high temporal resolution, significantly enhance visual saliency maps. However, the composite noise inherent to spike camera imaging introduces discontinuities in saliency detection. Low-quality samples further distort model predictions, leading to saliency bias. To address these challenges, we propose Spike-navigated Optimal TrAnsport Saliency Region Detection (SOTA), a framework that leverages the strengths of spike cameras while mitigating biases in both spatial and temporal dimensions. Our method introduces Spike-based Micro-debias (SM) to capture subtle frame-to-frame variations and preserve critical details, even under minimal scene or lighting changes. Additionally, Spike-based Global-debias (SG) refines predictions by reducing inconsistencies across diverse conditions. Extensive experiments on real and synthetic datasets demonstrate that SOTA outperforms existing methods by eliminating composite noise bias. Our code and dataset will be released at https://github.com/lwxfight/sota.
format Preprint
id arxiv_https___arxiv_org_abs_2505_00394
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SOTA: Spike-Navigated Optimal TrAnsport Saliency Region Detection in Composite-bias Videos
Liu, Wenxuan
Deng, Yao
Chen, Kang
Zhong, Xian
Yu, Zhaofei
Huang, Tiejun
Computer Vision and Pattern Recognition
Existing saliency detection methods struggle in real-world scenarios due to motion blur and occlusions. In contrast, spike cameras, with their high temporal resolution, significantly enhance visual saliency maps. However, the composite noise inherent to spike camera imaging introduces discontinuities in saliency detection. Low-quality samples further distort model predictions, leading to saliency bias. To address these challenges, we propose Spike-navigated Optimal TrAnsport Saliency Region Detection (SOTA), a framework that leverages the strengths of spike cameras while mitigating biases in both spatial and temporal dimensions. Our method introduces Spike-based Micro-debias (SM) to capture subtle frame-to-frame variations and preserve critical details, even under minimal scene or lighting changes. Additionally, Spike-based Global-debias (SG) refines predictions by reducing inconsistencies across diverse conditions. Extensive experiments on real and synthetic datasets demonstrate that SOTA outperforms existing methods by eliminating composite noise bias. Our code and dataset will be released at https://github.com/lwxfight/sota.
title SOTA: Spike-Navigated Optimal TrAnsport Saliency Region Detection in Composite-bias Videos
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.00394