Similarity-Aware Selective State-Space Modeling for Semantic Correspondence

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Kim, Seungwook, Cho, Minsu
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866918150420824064
author Kim, Seungwook
Cho, Minsu
author_facet Kim, Seungwook
Cho, Minsu
contents Establishing semantic correspondences between images is a fundamental yet challenging task in computer vision. Traditional feature-metric methods enhance visual features but may miss complex inter-correlation relationships, while recent correlation-metric approaches are hindered by high computational costs due to processing 4D correlation maps. We introduce MambaMatcher, a novel method that overcomes these limitations by efficiently modeling high-dimensional correlations using selective state-space models (SSMs). By implementing a similarity-aware selective scan mechanism adapted from Mamba's linear-complexity algorithm, MambaMatcher refines the 4D correlation map effectively without compromising feature map resolution or receptive field. Experiments on standard semantic correspondence benchmarks demonstrate that MambaMatcher achieves state-of-the-art performance.
format Preprint
id arxiv_https___arxiv_org_abs_2509_24318
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Similarity-Aware Selective State-Space Modeling for Semantic Correspondence
Kim, Seungwook
Cho, Minsu
Computer Vision and Pattern Recognition
Establishing semantic correspondences between images is a fundamental yet challenging task in computer vision. Traditional feature-metric methods enhance visual features but may miss complex inter-correlation relationships, while recent correlation-metric approaches are hindered by high computational costs due to processing 4D correlation maps. We introduce MambaMatcher, a novel method that overcomes these limitations by efficiently modeling high-dimensional correlations using selective state-space models (SSMs). By implementing a similarity-aware selective scan mechanism adapted from Mamba's linear-complexity algorithm, MambaMatcher refines the 4D correlation map effectively without compromising feature map resolution or receptive field. Experiments on standard semantic correspondence benchmarks demonstrate that MambaMatcher achieves state-of-the-art performance.
title Similarity-Aware Selective State-Space Modeling for Semantic Correspondence
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.24318