Open-source tools developed as part of the AWARE project.
Three surprising results from evaluating 21 augmentation strategies on adverse weather segmentation.
Training data diversity yields 3.1× larger improvements than any single augmentation strategy. Adding diverse real-world datasets provides +7.08 pp mIoU, while the best augmentation adds +2.27 pp.
Methods that produce visually better images (lower FID, higher SSIM) actually yield worse downstream performance. Aggressive transformations that distort appearance improve robustness more than photorealistic synthesis.
Synthetic and real adverse weather data compete rather than complement. When training data already includes real adverse conditions (Stage 2), augmentation gains nearly vanish, falling from +2.27 to +0.65 pp.
Interactive overview of the complete AWARE evaluation pipeline — from data curation through generative augmentation to downstream evaluation. Hover over any component for details, click to trace its full data path.
Dive deeper into the AWARE results with dedicated interactive pages.
Browse generation results across all 21 augmentation strategies. Filter by family and strategy, and compare outputs side-by-side.
Interactive charts, per-domain and per-model breakdowns, supplementary figures, and downloadable CSV data referenced in the paper.
Complete collection of CLIP text prompts used by SWIFT for two-stage weather classification, including fog counter-prompts.
Interactive comparison of baseline model predictions across datasets and weather conditions, with slider and overlay comparison tools.
Adverse Weather Augmentation Collection from Segment Anything 1 Billion — 71,400 balanced images across 7 weather conditions.
Each stripe shows the average color distribution of images classified into that weather condition, revealing distinct visual signatures.
Stage 1 performance ranking. Click column headers to sort. All values are mIoU (mean Intersection over Union).
Loading leaderboard data...
If you find this work useful, please cite our paper.
@article{aware2025,
title={Enhancing Computer Vision Robustness in Adverse Weather
Conditions through Advanced Data Augmentation},
author={[Author Name]},
journal={IEEE Access},
year={2025},
volume={},
pages={},
doi={}
}