FSMODNet: A Closer Look at Few-Shot Detection in Multispectral Data

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Nkegoum, Manuel, Pham, Minh-Tan, Fromont, Élisa, Avignon, Bruno, Lefèvre, Sébastien
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908558405140480
author Nkegoum, Manuel
Pham, Minh-Tan
Fromont, Élisa
Avignon, Bruno
Lefèvre, Sébastien
author_facet Nkegoum, Manuel
Pham, Minh-Tan
Fromont, Élisa
Avignon, Bruno
Lefèvre, Sébastien
contents Few-shot multispectral object detection (FSMOD) addresses the challenge of detecting objects across visible and thermal modalities with minimal annotated data. In this paper, we explore this complex task and introduce a framework named "FSMODNet" that leverages cross-modality feature integration to improve detection performance even with limited labels. By effectively combining the unique strengths of visible and thermal imagery using deformable attention, the proposed method demonstrates robust adaptability in complex illumination and environmental conditions. Experimental results on two public datasets show effective object detection performance in challenging low-data regimes, outperforming several baselines we established from state-of-the-art models. All code, models, and experimental data splits can be found at https://anonymous.4open.science/r/Test-B48D.
format Preprint
id arxiv_https___arxiv_org_abs_2509_20905
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FSMODNet: A Closer Look at Few-Shot Detection in Multispectral Data
Nkegoum, Manuel
Pham, Minh-Tan
Fromont, Élisa
Avignon, Bruno
Lefèvre, Sébastien
Computer Vision and Pattern Recognition
Few-shot multispectral object detection (FSMOD) addresses the challenge of detecting objects across visible and thermal modalities with minimal annotated data. In this paper, we explore this complex task and introduce a framework named "FSMODNet" that leverages cross-modality feature integration to improve detection performance even with limited labels. By effectively combining the unique strengths of visible and thermal imagery using deformable attention, the proposed method demonstrates robust adaptability in complex illumination and environmental conditions. Experimental results on two public datasets show effective object detection performance in challenging low-data regimes, outperforming several baselines we established from state-of-the-art models. All code, models, and experimental data splits can be found at https://anonymous.4open.science/r/Test-B48D.
title FSMODNet: A Closer Look at Few-Shot Detection in Multispectral Data
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.20905