Conference paper Open Access
ConfMix: Unsupervised Domain Adaptation for Object Detection via Confidence-based Mixing
Giulio Mattolin; Luca Zanella; Elisa Ricci; Yiming Wang
Unsupervised Domain Adaptation (UDA) for object detection aims to adapt a model trained on a source domain to detect instances from a new target domain for which annotations are not available. Different from traditional approaches, we propose ConfMix, the first method that introduces a sample mixing strategy based on region-level detec- tion confidence for adaptive object detector learning. We mix the local region of the target sample that corresponds to the most confident pseudo detections with a source image, and apply an additional consistency loss term to gradually adapt towards the target data distribution. In order to robustly define a confidence score for a region, we exploit the confidence score per pseudo detection that accounts for both the detector-dependent confidence and the bounding box uncertainty. Moreover, we propose a novel pseudo la- belling scheme that progressively filters the pseudo target detections using the confidence metric that varies from a loose to strict manner along the training. We perform ex- tensive experiments with three datasets, achieving state-of-the-art performance in two of them and approaching the supervised target model performance in the other.
| Name | Size | |
|---|---|---|
|
2023_WACV_Giulio_DA_ObjectDection.pdf
md5:581e64f7a0c37104b51ee2466494f0b9 |
5.7 MB | Download |
| All versions | This version | |
|---|---|---|
| Views | 191 | 191 |
| Downloads | 63 | 63 |
| Data volume | 357.8 MB | 357.8 MB |
| Unique views | 180 | 180 |
| Unique downloads | 60 | 60 |
- Publication date:
- November 17, 2022
- DOI:
-
- Keyword(s):
- Unsupervised Domain Adaptation object detection sample mixing region-level detection confidence object detector learning
- Grants:
-
European Commission:
- MARVEL - Multimodal Extreme Scale Data Analytics for Smart Cities Environments (957337)
- Meeting:
- EEE/CVF Winter Conference on Applications of Computer Vision 2023 (WACV 2023), Waikoloa, Hawaii, 3-7 January 2023
- Related identifiers:
- Supplementary material
- https://github.com/giuliomattolin/ConfMix (Software)
- https://www.cityscapes-dataset.com/ (Dataset)
- https://fcav.engin.umich.edu/projects/driving-in-the-matrix (Dataset)
- https://www.cvlibs.net/datasets/kitti/ (Dataset)
- Communities:
- License (for files):
- Creative Commons Attribution 4.0 International