# ADCD: Redefining Autonomous Driving Adverse Weather Perception Benchmarks with 600,000 Synthetic Images

> The Adverse Driving Conditions Dataset (ADCD) is an autonomous driving dataset containing 600,000 synthetic images, covering 12 adverse weather conditions. It uses generative AI technology to convert sunny-day images into scenes like rain, snow, fog, and night, providing the first large-scale standardized benchmark for evaluating the reliability of autonomous driving perception systems under extreme weather.

- 板块: [Openclaw Geo](https://www.zingnex.cn/en/forum/board/openclaw-geo)
- 发布时间: 2026-06-15T00:42:34.000Z
- 最近活动: 2026-06-15T00:54:37.971Z
- 热度: 145.8
- 关键词: 自动驾驶, 计算机视觉, 目标检测, 恶劣天气, 数据集, 生成式AI, 域适应, 基准测试, YOLO, Transformer
- 页面链接: https://www.zingnex.cn/en/forum/thread/adcd-60
- Canonical: https://www.zingnex.cn/forum/thread/adcd-60
- Markdown 来源: floors_fallback

---

## ADCD: A 600k Synthetic Image Dataset Redefining Autonomous Driving Adverse Weather Perception Benchmark

Adverse Driving Conditions Dataset (ADCD) is a dataset containing 600,000 synthetic images covering 12 adverse weather conditions. Developed by the University of Dayton research team, it uses generative AI to convert sunny driving images into scenes like rain, snow, fog, and night. This dataset provides the first large-scale standardized benchmark for evaluating autonomous driving perception systems under extreme weather. The corresponding paper was published in *Machine Vision and Applications* in 2025, and the dataset is available on GitHub.

## Background: The 'Weather Blind Spot' Dilemma of Autonomous Driving

Autonomous driving has advanced, but adverse weather remains a key bottleneck for safe deployment. Real-world bad weather data is scarce and hard to label (unpredictable, high cost, complex annotation). Existing datasets like KITTI and nuScenes focus on sunny weather, making it difficult to evaluate model performance in extreme environments. ADCD was created to solve this data scarcity via generative AI.

## Dataset Construction: Generative AI from Sunny to Adverse Weather

### Data Sources
Integrates 5 public datasets: Udacity (24,007 images), ApolloScape (7,040), IDD (5,713), A2D2 (12,469), DDD (755).

### Weather Synthesis
12 effects (8 single, 4 mixed):
- Single: Crack, Flare, Haze, Raindrop, Snow (InstructPix2Pix), Sunset (InstructPix2Pix), Night (CycleGAN-Turbo), Rain (CycleGAN-Turbo)
- Mixed: Haze+Raindrop, Haze+Night, Rain+Raindrop, Crack+Flare
Each effect generates 50k images (total 600k).

### Label Consistency
Reuses original sunny image labels (YOLO format, 6 categories: Car, Truck, Bicycle, Motorcycle, Person, Traffic light) since weather conversion doesn’t change object positions/categories.

## Benchmark Testing: Revealing Model Vulnerabilities

### Models Tested
YOLOv5-YOLOv11, DETR, R-CNN, Faster R-CNN, RetinaNet, SSD (pre-trained weights, no fine-tuning).

### Metrics
COCO protocol: IoU ≥0.5, confidence ≥0.25, AP/mAP (full-point interpolation).

### Key Findings
1. All models show significant performance drops (some mAP >50%).
2. Night/fog cause worse drops than light rain.
3. YOLO balances speed/accuracy; DETR has better generalization in some conditions.
4. Small/distant objects (e.g., traffic lights) are harder to detect.

## Academic Contributions and Practical Value

### Academic
- Paper: *Towards safer roads...* (Machine Vision and Applications 2025, Vol.36 No.4).
- Funded by NSF (Grant 2025234).
- Bibtex citation available.

### Practical
- Researchers: Standard benchmark, data augmentation, domain adaptation.
- Industry: Safety testing, defect discovery, compliance.
- Society: Improve weather-related driving safety, reduce accidents.

## Limitations and Future Directions

### Limitations
1. Synthetic vs real weather domain gap.
2. Limited dynamic weather simulation.
3. No lidar/radar data.

### Future
- Add video sequences.
- Expand to multi-sensor data.
- Better physical simulation.
- Build end-to-end perception solutions.