Datasets › ACDC (Adverse Conditions Dataset with Correspondences)

ACDC (Adverse Conditions Dataset with Correspondences)

archive 2025-07-28

We introduce ACDC, the Adverse Conditions Dataset with Correspondences for training and testing semantic segmentation methods on adverse visual conditions. It comprises a large set of 4006 images which are evenly distributed between fog, nighttime, rain, and snow. Each adverse-condition image comes with a high-quality fine pixel-level semantic annotation, a corresponding image of the same scene taken under normal conditions and a binary mask that distinguishes between intra-image regions of clear and uncertain semantic content.

ACDC supports two tasks: 1. standard semantic segmentation 2. uncertainty-aware semantic segmentation

Benchmarks archive 2025-07-28

All 5 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

First row (archive order)PaperCode
Domain Adaptation Cityscapes to ACDC SoRA mIoU 78.8 SoRA: Singular Value Decomposed Low-Rank Adaptation for... ysj9909/DG-SoRA 16 Compare
Source-Free Domain Adaptation Cityscapes to ACDC CMA mIoU 69.1 Contrastive Model Adaptation for Cross-Condition... brdav/cma 2 Compare
Domain Generalization Cityscapes to ACDC ADSI mIoU 70.21 Domain Generalization through Attenuation of... ReijiSoftmaxSaito/ADSI 1 Compare
Foggy Scene Segmentation ACDC (Adverse Conditions Dataset with Correspondences) BWG mIoU 76.7 Learning Generalized Segmentation for Foggy-scenes by... BiQiWHU/BWG 1 Compare
Unsupervised Semantic Segmentation ACDC (Adverse Conditions Dataset with Correspondences) Segmenter ViT-S/16 mIoU 16.7 Drive&Segment: Unsupervised Semantic Segmentation of... vobecant/DriveAndSegment 1 Compare

Papers archive 2025-07-28

18 shown of 18 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 31. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

DateSamples run Syntology
Domain Generalization through Attenuation of Domain-Specific Information 1 1 9 Apr 2025 not harvested
SoRA: Singular Value Decomposed Low-Rank Adaptation for Domain Generalizable Representation Learning 1 1 5 Dec 2024 not harvested
CoDA: Instructive Chain-of-Domain Adaptation with Severity-Aware Visual Prompt Tuning 1 1 26 Mar 2024 not harvested
Learning Generalized Segmentation for Foggy-scenes by Bi-directional Wavelet Guidance 1 1 28 Feb 2024 not harvested
Stronger, Fewer, & Superior: Harnessing Vision Foundation Models for Domain Generalized Semantic Segmentation 2 1 7 Dec 2023 not harvested
Learning Content-enhanced Mask Transformer for Domain Generalized Urban-Scene Segmentation 1 2 1 Jul 2023 not harvested
Hyperbolic Active Learning for Semantic Segmentation under Domain Shift 1 1 19 Jun 2023 ran 3 of 8 samples (5 unverified)
Contrastive Model Adaptation for Cross-Condition Robustness in Semantic Segmentation 1 1 9 Mar 2023 ran 5 of 6 samples (1 unverified)
MIC: Masked Image Consistency for Context-Enhanced Domain Adaptation 1 1 2 Dec 2022 not harvested
VBLC: Visibility Boosting and Logit-Constraint Learning for Domain Adaptive Semantic Segmentation under Adverse Conditions 1 2 22 Nov 2022 not harvested
Refign: Align and Refine for Adaptation of Semantic Segmentation to Adverse Conditions 1 2 14 Jul 2022 ran 0 of 7 samples (7 unverified)
HRDA: Context-Aware High-Resolution Domain-Adaptive Semantic Segmentation 1 1 27 Apr 2022 not harvested
Drive&Segment: Unsupervised Semantic Segmentation of Urban Scenes via Cross-modal Distillation 1 1 21 Mar 2022 ran 1 of 2 samples (1 unverified; 2 pointer-only for licence)
DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation 3 1 29 Nov 2021 ran 6 of 6 samples (0 unverified; 6 pointer-only for licence)
DANNet: A One-Stage Domain Adaptation Network for Unsupervised Nighttime Semantic Segmentation 1 1 22 Apr 2021 ran 3 of 4 samples (1 unverified)
DACS: Domain Adaptation via Cross-domain Mixed Sampling 2 1 17 Jul 2020 ran 3 of 4 samples (1 unverified; 3 pointer-only for licence)
Map-Guided Curriculum Domain Adaptation and Uncertainty-Aware Evaluation for Semantic Nighttime Image Segmentation 1 1 28 May 2020 not harvested
FDA: Fourier Domain Adaptation for Semantic Segmentation 3 1 11 Apr 2020 ran 2 of 3 samples (1 unverified; 3 pointer-only for licence)

Dataset loaders archive 2025-07-28

No loader listed in the archive.

Tasks archive 2025-07-28

License archive 2025-07-28

No licence recorded in the archive. Absence here is not a statement about the dataset's terms.

Modalities archive 2025-07-28

No modality tagged.

Languages archive 2025-07-28

No language tagged.

Variants archive 2025-07-28

  • ACDC (Adverse Conditions Dataset with Correspondences)
  • Cityscapes to ACDC

2 variant names, as the archive lists them.

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