---
title: 'ACDC: Dataset for Adverse Conditions'
url: https://www.emergentmind.com/topics/adverse-conditions-dataset-with-correspondences-acdc
type: topic
---

# ACDC: Dataset for Adverse Conditions

The Adverse Conditions Dataset with Correspondences (ACDC) is a large-scale benchmark for semantic driving scene perception under adverse visual conditions, introduced to address the dominance of daytime, clear-weather imagery in prior autonomous-driving datasets and the consequent fragility of perception systems outside nominal conditions [2104.13395]. It consists of 8012 images, including 4006 adverse-condition images equally distributed between fog, nighttime, rain, and snow, together with corresponding normal-condition images of the same scenes and privileged annotations designed to expose both semantic content and irreducible ambiguity under degraded visibility [2104.13395]. ACDC was designed for robust semantic segmentation, object detection, instance segmentation, panoptic segmentation, and uncertainty-aware semantic segmentation, and it has become a central benchmark for supervised learning, unsupervised domain adaptation, source-free adaptation, continual adaptation, and augmentation-based robustness studies in adverse-weather perception [2104.13395], [2207.06825], [2303.05194], [2407.13437], [2603.24322].

## 1. Motivation and problem setting

ACDC was introduced from the premise that level-5 driving automation requires a robust visual perception system that can parse input images under any condition, whereas existing driving datasets for dense semantic perception are either dominated by images captured under normal conditions or are small in scale [2104.13395]. Large-scale semantic-driving datasets such as Cityscapes, Mapillary Vistas, and KITTI have very limited coverage of adverse conditions, while datasets covering adverse weather such as Oxford RobotCar and BDD100K either lack dense pixel-level annotations or provide only limited and often unreliable adverse-condition annotations [2104.13395].

A second motivation was that existing pixel-level annotations for adverse conditions generally did not account for aleatory uncertainty: some image regions are intrinsically ambiguous because of darkness, sensor noise, fog, snowfall, lens artifacts, or related degradations [2104.13395]. ACDC therefore coupled dense semantic annotation with an explicit notion of invalidity, so that ambiguous regions could be separated from regions whose semantics remain visually ascertainable [2104.13395]. This design choice also underpins the dataset’s uncertainty-aware segmentation task.

A third motivation was methodological rather than purely curatorial. ACDC was constructed with corresponding normal-condition images of the same scenes, enabling adaptation and cross-condition transfer methods to exploit weak supervision from easier views of the same locations [2104.13395]. This property later became central to alignment- and contrast-based adaptation methods such as Refign and CMA, which explicitly use the reference-target pairing provided by ACDC [2207.06825], [2303.05194].

## 2. Composition, acquisition, and scene taxonomy

ACDC consists of a large set of 8012 images, half of which, 4006, are equally distributed between four common adverse conditions: fog, nighttime, rain, and snow [2104.13395]. The adverse subset contains 1000 fog images, 1006 nighttime images, 1000 rain images, and 1000 snow images [2104.13395]. Each adverse-condition image comes with a corresponding image of the same scene under normal conditions, and 1503 of the corresponding normal-condition images feature panoptic annotations, raising the total number of annotated images to 5509 [2104.13395].

The dataset uses the same 19 semantic classes as Cityscapes, enabling direct transfer and comparison in semantic segmentation and domain adaptation settings [2104.13395], [2211.12256]. For instance- and panoptic-level tasks, ACDC provides annotations for “things” classes such as vehicles and humans, alongside “stuff” classes in standard panoptic form [2104.13395].

The following summary organizes the core dataset statistics reported for ACDC.

| Component | Reported content |
|---|---|
| Total images | 8012 |
| Adverse images | 4006 |
| Adverse conditions | Fog, nighttime, rain, snow |
| Annotated normal-condition images | 1503 |
| Total annotated images | 5509 |
| Semantic classes | 19, matching Cityscapes |

Data recording was carried out with a GoPro Hero 5 at 1080p, 8-bit RGB, and 30 Hz [2104.13395]. The adverse subsets were recorded so that each acquisition emphasizes a single adversity: fog was recorded under foggy daytime, nighttime strictly at night, rain in daylight with clear rain effects, and snow with both falling snow and snow cover on the ground [2104.13395]. This strict visual separation increases domain shift and makes the benchmark deliberately difficult [2104.13395].

All adverse conditions and their reference images are split into train, validation, and test according to a strict geographic split with zero overlap, and test labels are withheld for benchmarking [2104.13395]. This geographic separation is important because it limits trivial memorization of locations and preserves the benchmark’s role as a cross-condition generalization testbed.

## 3. Annotation protocol and uncertainty modeling

A defining feature of ACDC is its annotation protocol for adverse imagery. Each adverse-condition image has high-quality pixel-level panoptic annotation, but the annotation is produced in a two-stage process specifically intended to separate visible semantics from ambiguity [2104.13395].

In stage 1, a human annotator labels what can be unambiguously determined from the adverse image alone [2104.13395]. In stage 2, the annotator cross-references the corresponding normal-condition image and video, revisiting ambiguous regions; if a region can now be assigned a correct label using this privileged information, it is labeled, whereas regions that remain indiscernible stay unlabeled and are excluded [2104.13395]. Each adverse-condition annotation therefore includes a binary mask, denoted \(J\), indicating invalid regions: pixels whose label changes between stage 1 and stage 2 are marked invalid, while the remainder are treated as valid [2104.13395].

This protocol yields a benchmark in which not all pixels are equally observable. Reported annotation statistics indicate that about 85% of pixels in adverse images are labeled as valid, 7.5% are labeled as invalid after the privileged step, and the remainder are unlabelable [2104.13395]. ACDC therefore differs from conventional semantic segmentation datasets in that uncertainty is not treated merely as model error; part of it is embedded in the ground-truth specification itself.

That design supports a distinct task: uncertainty-aware semantic segmentation [2104.13395]. In this setting, models predict both semantic labels and a confidence map per pixel, and evaluation uses Average Uncertainty-aware IoU (AUIoU), which averages UiOU over confidence thresholds while rewarding systems that assign low confidence to truly invalid regions and penalizing overconfident predictions in ambiguous areas [2104.13395]. The benchmark thus formalizes a setting in which abstention can be correct.

A frequent misconception is that ACDC’s “correspondences” imply dense pixel-wise correspondence labels between normal and adverse images. The dataset does not provide pixel- or instance-level correspondences; rather, it provides image-level correspondences of the same scene and approximate viewpoint, obtained by synchronizing GPS logs from adverse and normal recordings with dynamic-programming-based sequence alignment [2104.13395], [2207.06825]. Later methods such as Refign and CMA estimate dense alignments on top of these image-level pairs rather than reading them directly from the dataset [2207.06825], [2303.05194].

## 4. Correspondences and supported perception tasks

The “with Correspondences” component of ACDC’s name refers to its paired normal/adverse images of the same scenes [2104.13395]. These correspondences were created by synchronizing GPS logs from adverse and normal recordings, producing approximate viewpoint matches that can serve both annotation and learning [2104.13395]. Their first role is curatorial: privileged normal-condition information improves labeling quality for adverse scenes. Their second role is algorithmic: they provide weak supervision for cross-condition adaptation, feature alignment, and robustness analysis [2104.13395], [2207.06825], [2303.05194].

ACDC supports five benchmark tasks: semantic segmentation, panoptic segmentation, instance segmentation, object detection, and uncertainty-aware semantic segmentation [2104.13395]. This breadth made it possible for the dataset to function as more than a segmentation benchmark. In later object-detection experiments, ACDC was used as a real-world test set for evaluating robustness under adverse weather, and because it does not directly provide bounding boxes for traffic lights and traffic signs, those labels were generated from provided semantic segmentation masks using the `skimage.measure.regionprops` Python function [2505.08228]. In that study, evaluation used 250 images for each adverse weather condition from ACDC, for a total of 1000 ACDC images [2505.08228].

The correspondence structure has also been central to semantic adaptation work. Refign treats ACDC reference-target pairs as weak supervision and augments self-training-based unsupervised domain adaptation with uncertainty-aware dense alignment and adaptive label correction [2207.06825]. CMA studies source-free model adaptation under the constraint that source data are inaccessible, using ACDC’s GPS-matched normal/adverse image pairs to learn condition-invariant features via contrastive learning [2303.05194]. FREST likewise operates in a source-free setting on ACDC, alternating condition-embedding learning and feature restoration using unlabeled adverse images paired with reference images under normal conditions from similar geolocations [2407.13437].

This suggests that ACDC’s principal scientific contribution is not only the adverse imagery itself but the coupling of adverse visual degradation with a weakly aligned, normal-condition companion domain. A plausible implication is that the dataset supports a broader class of adaptation paradigms than benchmarks that provide adverse images alone.

## 5. ACDC as a benchmark for robust segmentation and adaptation

ACDC rapidly became a standard target domain for semantic segmentation under adverse conditions. In “Semantic Segmentation under Adverse Conditions: A Weather and Nighttime-aware Synthetic Data-based Approach,” ACDC is used only for evaluation, with the validation split of 506 images spanning rainy, foggy, snowy, and nighttime conditions [2210.05626]. In that study, a DeepLabV3+-based model with weather-aware and time-aware supervisors achieved an overall ACDC mIoU of 0.49, compared with 0.35 for the DeepLabV3+ baseline, while maintaining 0.75 mIoU on Cityscapes [2210.05626]. The paper reports that baseline models trained on Cityscapes only perform poorly on ACDC, with overall mIoU in the range 0.35–0.42, and that naive fine-tuning on synthetic adverse weather data improves ACDC but can substantially reduce Cityscapes performance [2210.05626].

WeatherProof re-examines ACDC from the perspective of language-guided segmentation. Using InternImage and CLIP-derived weather composition “side information,” it reports validation-set results on rain, fog, and snow, with average mIoU increasing from 75.3 for InternImage to 81.0 for InternImage + the proposed method, and an mIoU increase from 76.17 to 82.6 in another reported ACDC setting [2403.14874]. The study emphasizes that although ACDC divides data by primary weather effect, real-world images can contain multiple overlapping degradations, and that language-guided weather composition modeling is particularly effective in such mix scenarios [2403.14874].

In unsupervised domain adaptation, ACDC has functioned as the canonical adverse-weather target for Cityscapes-to-adverse transfer. VBLC evaluates Cityscapes \(\rightarrow\) ACDC on the official test set and reports 47.8 mIoU with a DeepLab-v2 backbone versus 45.7 for FDA, and 64.2 mIoU with a SegFormer backbone versus 55.3 for DAFormer [2211.12256]. Heuristic Self-Paced Learning later reports state-of-the-art Cityscapes \(\rightarrow\) ACDC performance of 72.9 mIoU on the ACDC test set and 72.7 on the validation set when built on HRDA [2603.24322].

In source-free adaptation, ACDC’s correspondences are especially valuable because they compensate, in part, for the absence of source data during adaptation. CMA reports 69.1 mIoU on the ACDC test set with a SegFormer backbone, improving over source-free baselines such as TENT, HCL, URMA, and URMA+SimT [2303.05194]. FREST reports 70.7 mIoU on Cityscapes \(\rightarrow\) ACDC, improving over CMA’s 69.1, and also reports per-condition gains on fog, night, rain, and snow [2407.13437].

ACDC has also been used to study continual multi-target adaptation across adverse conditions rather than one-shot adaptation to a single target. In “Semantic Segmentation in Multiple Adverse Weather Conditions with Domain Knowledge Retention,” the four ACDC splits are treated as sequential target domains in the order Cityscapes \(\rightarrow\) Night \(\rightarrow\) Rain \(\rightarrow\) Fog \(\rightarrow\) Snow [2401.07459]. With a DAFormer backbone, the proposed method reports an averaged mIoU of 65.7 and accumulated forgetting of 3.6, compared with 60.1 and 11.3 for MIC [2401.07459]. This use of ACDC highlights that the dataset can be decomposed into weather-specific target domains for studying knowledge retention and catastrophic forgetting.

## 6. Empirical difficulty, error modes, and broader impact

ACDC is widely regarded as exceptionally challenging because adverse conditions induce severe visibility degradation, large domain shifts, and condition-specific artifacts [2505.08228], [2210.05626]. Reported causes include rain droplets, snow flakes, fog acting as a low-pass filter, low light at night, lens flare, altered object appearance, snow accumulation, and wet or reflective surfaces [2210.05626]. These degradations substantially reduce performance for models trained only under standard conditions.

The original empirical study already showed that performance on ACDC drops markedly relative to clear-weather benchmarks and that nighttime is generally the hardest domain [2104.13395]. The same study found that many unsupervised domain adaptation methods effective in synthetic-to-real transfer brought little or no gain over source-only baselines on ACDC, underscoring that real adverse weather is a harder setting than stylized or synthetic shifts [2104.13395]. It also reported that models trained on all conditions, described as “uber” models, outperform ensembles specialized to single conditions, and that nighttime-trained models generalize better to other conditions than vice versa [2104.13395].

Subsequent work reinforces these conclusions across tasks. In object detection, ACDC serves as a real-world test set demonstrating a pronounced drop in detection performance under adverse weather when detectors are trained only on clear-weather data; data augmentation with Instruct Pix2Pix improved Faster R-CNN robustness in fog, night, rain, and snow, with especially visible gains at night and for pedestrian detection [2505.08228]. In one reported night setting, AP50 for the walker class increased from 0.415 to 0.500 for Faster R-CNN when augmented data were used [2505.08228].

ACDC has also become a reference point in discussions of what dataset correspondence means across modalities. A 2026 LiDAR adaptation paper contrasts ACDC with a new LiDAR dataset suite, noting that ACDC offers real-world adverse-weather data but not frame-to-frame paired clear/adverse LiDAR scans with exact physical correspondence [2604.10213]. This comparison does not diminish ACDC’s role; rather, it clarifies that ACDC’s strength lies in RGB semantic perception with image-level scene correspondence, not in sensor-level one-to-one physical re-renderings.

The dataset’s long-term impact is methodological. Refign uses ACDC correspondences for adaptive pseudo-label correction [2207.06825]. CMA uses them for cross-domain contrastive learning when source data are inaccessible [2303.05194]. FREST uses them to restore adverse features toward normal-condition embeddings [2407.13437]. WeatherProof uses ACDC to test language-guided side information for segmentation under overlapping weather effects [2403.14874]. HeuSCM uses ACDC as one of three widely used benchmarks to validate a learned curriculum for class scheduling in domain-adaptive segmentation [2603.24322]. Across these works, ACDC functions simultaneously as a dataset, a benchmark, and a formalization of adverse-condition perception in which uncertainty, correspondence, and domain shift are all first-class objects.

A plausible implication is that ACDC’s enduring importance stems from the combination of three properties rarely found together: real adverse-weather imagery at scale, dense semantic annotation with explicit invalid-region modeling, and cross-condition image-level correspondences. That combination has made it a persistent testbed not only for measuring robustness, but for defining what robustness should mean in semantic driving scene perception under genuinely adverse visual conditions.

Source: https://www.emergentmind.com/topics/adverse-conditions-dataset-with-correspondences-acdc