# Exploiting Low-confidence Pseudo-labels for Source-free Object Detection

arXiv: https://arxiv.org/abs/2310.12705 · HTML: https://mlsanity.com/paper/2310.12705/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 7 references that are in the map.
- First posted on arXiv 19 Oct 2023.

## Builds on (most cited first, 4 of 4)

- [FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling](https://mlsanity.com/paper/2110.08263/) (2021)
- [Semantic Foggy Scene Understanding with Synthetic Data](https://mlsanity.com/paper/1708.07819/) (2017)
- [A Free Lunch for Unsupervised Domain Adaptive Object Detection without Source Data](https://mlsanity.com/paper/2012.05400/) (2020)
- [Exploring Sequence Feature Alignment for Domain Adaptive Detection Transformers](https://mlsanity.com/paper/2107.12636/) (2021)

## Built on by (most cited first, 13 of 13)

- [Simplifying Source-Free Domain Adaptation for Object Detection: Effective Self-Training Strategies and Performance Insights](https://mlsanity.com/paper/2407.07586/) (2024)
- [Enhancing Source-Free Domain Adaptive Object Detection with Low-confidence Pseudo Label Distillation](https://mlsanity.com/paper/2407.13524/) (2024)
- [Source-free Domain Adaptive Object Detection in Remote Sensing Images](https://mlsanity.com/paper/2401.17916/) (2024)
- [Continual Test-Time Adaptation for Object Detection with Adaptive Monitoring and Randomized Restoration](https://mlsanity.com/paper/2406.16439/) (2024)
- [Source-Free Object Detection with Detection Transformer](https://mlsanity.com/paper/2510.11090/) (2025)
- [Source-free Domain Adaptation for Video Object Detection Under Adverse Image Conditions](https://mlsanity.com/paper/2404.15252/) (2024)
- [SFDLA: Source-Free Document Layout Analysis](https://mlsanity.com/paper/2503.18742/) (2025)
- [ViLAaD: Enhancing "Attracting and Dispersing'' Source-Free Domain Adaptation with Vision-and-Language Model](https://mlsanity.com/paper/2503.23529/) (2025)
- [Beyond Boundaries: Leveraging Vision Foundation Models for Source-Free Object Detection](https://mlsanity.com/paper/2511.07301/) (2025)
- [Source-Free Domain Adaptive Object Detection with Semantics Compensation](https://mlsanity.com/paper/2410.05557/) (2024)
- [Leveraging Confident Image Regions for Source-Free Domain-Adaptive Object Detection](https://mlsanity.com/paper/2501.10081/) (2025)
- [Towards Unbiased Source-Free Object Detection via Vision Foundation Models](https://mlsanity.com/paper/2601.12765/) (2026)
- [Simple Supervision Is Hard to Beat: A Bitter Lesson from Sparse Target Labels in Domain-Adaptive Object Detection](https://mlsanity.com/paper/2606.30795/) (2026)

## Abstract (arXiv)

Source-free object detection (SFOD) aims to adapt a source-trained detector to an unlabeled target domain without access to the labeled source data. Current SFOD methods utilize a threshold-based pseudo-label approach in the adaptation phase, which is typically limited to high-confidence pseudo-labels and results in a loss of information. To address this issue, we propose a new approach to take full advantage of pseudo-labels by introducing high and low confidence thresholds. Specifically, the pseudo-labels with confidence scores above the high threshold are used conventionally, while those between the low and high thresholds are exploited using the Low-confidence Pseudo-labels Utilization (LPU) module. The LPU module consists of Proposal Soft Training (PST) and Local Spatial Contrastive Learning (LSCL). PST generates soft labels of proposals for soft training, which can mitigate the label mismatch problem. LSCL exploits the local spatial relationship of proposals to improve the model's ability to differentiate between spatially adjacent proposals, thereby optimizing representational features further. Combining the two components overcomes the challenges faced by traditional methods in utilizing low-confidence pseudo-labels. Extensive experiments on five cross-domain object detection benchmarks demonstrate that our proposed method outperforms the previous SFOD methods, achieving state-of-the-art performance.