DOD-SA: Infrared-Visible Decoupled Object Detection with Single-Modality Annotations
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909736784363520 |
|---|---|
| author | Jin, Hang Gao, Chenqiang Guo, Junjie Liu, Fangcen Tian, Kanghui Chang, Qinyao |
| author_facet | Jin, Hang Gao, Chenqiang Guo, Junjie Liu, Fangcen Tian, Kanghui Chang, Qinyao |
| contents | Infrared-visible object detection has shown great potential in real-world applications, enabling robust all-day perception by leveraging the complementary information of infrared and visible images. However, existing methods typically require dual-modality annotations to output detection results for both modalities during prediction, which incurs high annotation costs. To address this challenge, we propose a novel infrared-visible Decoupled Object Detection framework with Single-modality Annotations, called DOD-SA. The architecture of DOD-SA is built upon a Single- and Dual-Modality Collaborative Teacher-Student Network (CoSD-TSNet), which consists of a single-modality branch (SM-Branch) and a dual-modality decoupled branch (DMD-Branch). The teacher model generates pseudo-labels for the unlabeled modality, simultaneously supporting the training of the student model. The collaborative design enables cross-modality knowledge transfer from the labeled modality to the unlabeled modality, and facilitates effective SM-to-DMD branch supervision. To further improve the decoupling ability of the model and the pseudo-label quality, we introduce a Progressive and Self-Tuning Training Strategy (PaST) that trains the model in three stages: (1) pretraining SM-Branch, (2) guiding the learning of DMD-Branch by SM-Branch, and (3) refining DMD-Branch. In addition, we design a Pseudo Label Assigner (PLA) to align and pair labels across modalities, explicitly addressing modality misalignment during training. Extensive experiments on the DroneVehicle dataset demonstrate that our method outperforms state-of-the-art (SOTA). |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_10445 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DOD-SA: Infrared-Visible Decoupled Object Detection with Single-Modality Annotations Jin, Hang Gao, Chenqiang Guo, Junjie Liu, Fangcen Tian, Kanghui Chang, Qinyao Computer Vision and Pattern Recognition 68T07, 68T45, 68U10 I.2.10 Infrared-visible object detection has shown great potential in real-world applications, enabling robust all-day perception by leveraging the complementary information of infrared and visible images. However, existing methods typically require dual-modality annotations to output detection results for both modalities during prediction, which incurs high annotation costs. To address this challenge, we propose a novel infrared-visible Decoupled Object Detection framework with Single-modality Annotations, called DOD-SA. The architecture of DOD-SA is built upon a Single- and Dual-Modality Collaborative Teacher-Student Network (CoSD-TSNet), which consists of a single-modality branch (SM-Branch) and a dual-modality decoupled branch (DMD-Branch). The teacher model generates pseudo-labels for the unlabeled modality, simultaneously supporting the training of the student model. The collaborative design enables cross-modality knowledge transfer from the labeled modality to the unlabeled modality, and facilitates effective SM-to-DMD branch supervision. To further improve the decoupling ability of the model and the pseudo-label quality, we introduce a Progressive and Self-Tuning Training Strategy (PaST) that trains the model in three stages: (1) pretraining SM-Branch, (2) guiding the learning of DMD-Branch by SM-Branch, and (3) refining DMD-Branch. In addition, we design a Pseudo Label Assigner (PLA) to align and pair labels across modalities, explicitly addressing modality misalignment during training. Extensive experiments on the DroneVehicle dataset demonstrate that our method outperforms state-of-the-art (SOTA). |
| title | DOD-SA: Infrared-Visible Decoupled Object Detection with Single-Modality Annotations |
| topic | Computer Vision and Pattern Recognition 68T07, 68T45, 68U10 I.2.10 |
| url | https://arxiv.org/abs/2508.10445 |