Papers › HRDA: Context-Aware High-Resolution Domain-Adaptive Semantic Segmentation

HRDA: Context-Aware High-Resolution Domain-Adaptive Semantic Segmentation

27 Apr 2022arXiv:2204.13132archive 2025-07-28

Lukas Hoyer, Dengxin Dai, Luc van Gool

Unsupervised domain adaptation (UDA) aims to adapt a model trained on the source domain (e.g. synthetic data) to the target domain (e.g. real-world data) without requiring further annotations on the target domain. This work focuses on UDA for semantic segmentation as real-world pixel-wise annotations are particularly expensive to acquire. As UDA methods for semantic segmentation are usually GPU memory intensive, most previous methods operate only on downscaled images. We question this design as low-resolution predictions often fail to preserve fine details. The alternative of training with random crops of high-resolution images alleviates this problem but falls short in capturing long-range, domain-robust context information. Therefore, we propose HRDA, a multi-resolution training approach for UDA, that combines the strengths of small high-resolution crops to preserve fine segmentation details and large low-resolution crops to capture long-range context dependencies with a learned scale attention, while maintaining a manageable GPU memory footprint. HRDA enables adapting small objects and preserving fine segmentation details. It significantly improves the state-of-the-art performance by 5.5 mIoU for GTA-to-Cityscapes and 4.9 mIoU for Synthia-to-Cityscapes, resulting in unprecedented 73.8 and 65.8 mIoU, respectively. The implementation is available at https://github.com/lhoyer/HRDA.

Paper PDF Code

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

lhoyer/hrda officialmentioned in papermentioned on GitHubpytorchNOASSERTION report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Domain Adaptation Image-to-Image Translation Segmentation Semantic Segmentation Synthetic-to-Real Translation Unsupervised Domain Adaptation Vocal Bursts Intensity Prediction

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

Task	Dataset	Model	Metric	Value	Rank at snapshot	Leaderboard	Report
Domain Adaptation	Cityscapes to ACDC	HRDA	mIoU	68.0	#7 of 16	Archive leaderboard	report
Domain Adaptation	GTA5 to Cityscapes	HRDA	mIoU	73.8	#6 of 28	Archive leaderboard	report
Domain Adaptation	SYNTHIA-to-Cityscapes	HRDA	mIoU	65.8	#7 of 33	Archive leaderboard	report
Image-to-Image Translation	GTAV-to-Cityscapes Labels	HRDA	mIoU	73.8	#3 of 22	Archive leaderboard	report
Image-to-Image Translation	SYNTHIA-to-Cityscapes	HRDA	mIoU (13 classes)	72.4	#3 of 28	Archive leaderboard	report
Semantic Segmentation	Dark Zurich	HRDA	mIoU	55.9	#5 of 14	Archive leaderboard	report
Semantic Segmentation	GTAV-to-Cityscapes Labels	HRDA	mIoU	73.8	#3 of 12	Archive leaderboard	report
Semantic Segmentation	SYNTHIA-to-Cityscapes	HRDA	Mean IoU	65.8	#3 of 7	Archive leaderboard	report
Synthetic-to-Real Translation	GTAV-to-Cityscapes Labels	HRDA	mIoU	73.8	#5 of 73	Archive leaderboard	report
Synthetic-to-Real Translation	SYNTHIA-to-Cityscapes	HRDA	MIoU (13 classes)	72.4	#5 of 38	Archive leaderboard	report
Synthetic-to-Real Translation	SYNTHIA-to-Cityscapes	HRDA	MIoU (16 classes)	65.8	#5 of 38	Archive leaderboard	report
Unsupervised Domain Adaptation	GTAV-to-Cityscapes Labels	HRDA	mIoU	73.8	#4 of 20	Archive leaderboard	report
Unsupervised Domain Adaptation	SYNTHIA-to-Cityscapes	HRDA	mIoU	65.8	#5 of 23	Archive leaderboard	report
Unsupervised Domain Adaptation	SYNTHIA-to-Cityscapes	HRDA	mIoU (13 classes)	72.4	#5 of 23	Archive leaderboard	report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections