DistillMatch: Leveraging Knowledge Distillation from Vision Foundation Model for Multimodal Image Matching

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Meng, Fan, Fan, Li, Zizhuo, Deng, Songchu, Ma, Yong, Ma, Jiayi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911163657224192
author Yang, Meng
Fan, Fan
Li, Zizhuo
Deng, Songchu
Ma, Yong
Ma, Jiayi
author_facet Yang, Meng
Fan, Fan
Li, Zizhuo
Deng, Songchu
Ma, Yong
Ma, Jiayi
contents Multimodal image matching seeks pixel-level correspondences between images of different modalities, crucial for cross-modal perception, fusion and analysis. However, the significant appearance differences between modalities make this task challenging. Due to the scarcity of high-quality annotated datasets, existing deep learning methods that extract modality-common features for matching perform poorly and lack adaptability to diverse scenarios. Vision Foundation Model (VFM), trained on large-scale data, yields generalizable and robust feature representations adapted to data and tasks of various modalities, including multimodal matching. Thus, we propose DistillMatch, a multimodal image matching method using knowledge distillation from VFM. DistillMatch employs knowledge distillation to build a lightweight student model that extracts high-level semantic features from VFM (including DINOv2 and DINOv3) to assist matching across modalities. To retain modality-specific information, it extracts and injects modality category information into the other modality's features, which enhances the model's understanding of cross-modal correlations. Furthermore, we design V2I-GAN to boost the model's generalization by translating visible to pseudo-infrared images for data augmentation. Experiments show that DistillMatch outperforms existing algorithms on public datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2509_16017
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DistillMatch: Leveraging Knowledge Distillation from Vision Foundation Model for Multimodal Image Matching
Yang, Meng
Fan, Fan
Li, Zizhuo
Deng, Songchu
Ma, Yong
Ma, Jiayi
Computer Vision and Pattern Recognition
I.4.3; I.5.2
Multimodal image matching seeks pixel-level correspondences between images of different modalities, crucial for cross-modal perception, fusion and analysis. However, the significant appearance differences between modalities make this task challenging. Due to the scarcity of high-quality annotated datasets, existing deep learning methods that extract modality-common features for matching perform poorly and lack adaptability to diverse scenarios. Vision Foundation Model (VFM), trained on large-scale data, yields generalizable and robust feature representations adapted to data and tasks of various modalities, including multimodal matching. Thus, we propose DistillMatch, a multimodal image matching method using knowledge distillation from VFM. DistillMatch employs knowledge distillation to build a lightweight student model that extracts high-level semantic features from VFM (including DINOv2 and DINOv3) to assist matching across modalities. To retain modality-specific information, it extracts and injects modality category information into the other modality's features, which enhances the model's understanding of cross-modal correlations. Furthermore, we design V2I-GAN to boost the model's generalization by translating visible to pseudo-infrared images for data augmentation. Experiments show that DistillMatch outperforms existing algorithms on public datasets.
title DistillMatch: Leveraging Knowledge Distillation from Vision Foundation Model for Multimodal Image Matching
topic Computer Vision and Pattern Recognition
I.4.3; I.5.2
url https://arxiv.org/abs/2509.16017