Source-Free Object Detection with Detection Transformer

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yao, Huizai, Zhao, Sicheng, Lu, Shuo, Chen, Hui, Li, Yangyang, Liu, Guoping, Xing, Tengfei, Yan, Chenggang, Tao, Jianhua, Ding, Guiguang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912644356636672
author Yao, Huizai
Zhao, Sicheng
Lu, Shuo
Chen, Hui
Li, Yangyang
Liu, Guoping
Xing, Tengfei
Yan, Chenggang
Tao, Jianhua
Ding, Guiguang
author_facet Yao, Huizai
Zhao, Sicheng
Lu, Shuo
Chen, Hui
Li, Yangyang
Liu, Guoping
Xing, Tengfei
Yan, Chenggang
Tao, Jianhua
Ding, Guiguang
contents Source-Free Object Detection (SFOD) enables knowledge transfer from a source domain to an unsupervised target domain for object detection without access to source data. Most existing SFOD approaches are either confined to conventional object detection (OD) models like Faster R-CNN or designed as general solutions without tailored adaptations for novel OD architectures, especially Detection Transformer (DETR). In this paper, we introduce Feature Reweighting ANd Contrastive Learning NetworK (FRANCK), a novel SFOD framework specifically designed to perform query-centric feature enhancement for DETRs. FRANCK comprises four key components: (1) an Objectness Score-based Sample Reweighting (OSSR) module that computes attention-based objectness scores on multi-scale encoder feature maps, reweighting the detection loss to emphasize less-recognized regions; (2) a Contrastive Learning with Matching-based Memory Bank (CMMB) module that integrates multi-level features into memory banks, enhancing class-wise contrastive learning; (3) an Uncertainty-weighted Query-fused Feature Distillation (UQFD) module that improves feature distillation through prediction quality reweighting and query feature fusion; and (4) an improved self-training pipeline with a Dynamic Teacher Updating Interval (DTUI) that optimizes pseudo-label quality. By leveraging these components, FRANCK effectively adapts a source-pre-trained DETR model to a target domain with enhanced robustness and generalization. Extensive experiments on several widely used benchmarks demonstrate that our method achieves state-of-the-art performance, highlighting its effectiveness and compatibility with DETR-based SFOD models.
format Preprint
id arxiv_https___arxiv_org_abs_2510_11090
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Source-Free Object Detection with Detection Transformer
Yao, Huizai
Zhao, Sicheng
Lu, Shuo
Chen, Hui
Li, Yangyang
Liu, Guoping
Xing, Tengfei
Yan, Chenggang
Tao, Jianhua
Ding, Guiguang
Computer Vision and Pattern Recognition
Artificial Intelligence
Source-Free Object Detection (SFOD) enables knowledge transfer from a source domain to an unsupervised target domain for object detection without access to source data. Most existing SFOD approaches are either confined to conventional object detection (OD) models like Faster R-CNN or designed as general solutions without tailored adaptations for novel OD architectures, especially Detection Transformer (DETR). In this paper, we introduce Feature Reweighting ANd Contrastive Learning NetworK (FRANCK), a novel SFOD framework specifically designed to perform query-centric feature enhancement for DETRs. FRANCK comprises four key components: (1) an Objectness Score-based Sample Reweighting (OSSR) module that computes attention-based objectness scores on multi-scale encoder feature maps, reweighting the detection loss to emphasize less-recognized regions; (2) a Contrastive Learning with Matching-based Memory Bank (CMMB) module that integrates multi-level features into memory banks, enhancing class-wise contrastive learning; (3) an Uncertainty-weighted Query-fused Feature Distillation (UQFD) module that improves feature distillation through prediction quality reweighting and query feature fusion; and (4) an improved self-training pipeline with a Dynamic Teacher Updating Interval (DTUI) that optimizes pseudo-label quality. By leveraging these components, FRANCK effectively adapts a source-pre-trained DETR model to a target domain with enhanced robustness and generalization. Extensive experiments on several widely used benchmarks demonstrate that our method achieves state-of-the-art performance, highlighting its effectiveness and compatibility with DETR-based SFOD models.
title Source-Free Object Detection with Detection Transformer
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2510.11090