adverse-weather

[ECCV 2024] Histoformer: Restoring Images in Adverse Weather Conditions via Histogram Transformer

[ICPR 2024 Best Paper]:"AllWeatherNet:Unified Image enhancement for autonomous driving under adverse weather and lowlight-conditions"

[ECCV‘24] Teaching Tailored to Talent: Adverse Weather Restoration via Prompt Pool and Depth-Anything Constraint

Python Package that makes Vehicle Dynamics Calculation for cars.

The survey on the computer vision works under the adverse weather conditions

LAWA: LiDAR Adverse Weather Augmentation method

Official companion repository of "Deep Image Restoration in Adverse Weather: A Survey" (Neural Networks 2027): methods, datasets, losses, metrics & benchmarks.

Adverse-weather image synthesis and YOLOv8-based object detection, developed for the SynthVision National Hackathon.

Agentic adverse-weather image restoration with Reliability-aware Action Control (RAC): degradation perception, reliability evaluation and EXECUTE/VERIFY/SUPPRESS action gating over specialist restoration tools.

Analyzing semantic segmentation robustness in adverse-weather driving scenes

Object detection for road users in adverse-weather conditions (fog, night, rain, snow) using YOLO11n on the ACDC dataset. Features a FastAPI inference service with Docker deployment, a Next.js web interface with visual bounding-box overlay, and a CustomTkinter desktop client. Trained across 8 classes.

SVOR: training-free snow removal for spinning LiDAR - 92.8 macro F1 on WADS at ~10 ms/frame on CPU, matching a RA-L 2023 method at ~1900x its speed. ROS-free C++17 core + ROS 2 node. Reference implementation: SnowClear.

Research on vehicle accident detection in adverse weather using YOLOv10 and VGG19 deep learning models.

Rainy-condition domain adaptation notes for semantic segmentation

Severity-aware adaptive inference for object detection in degraded visual environments. MSc thesis (2026) — physics-grounded routing framework for fog and rain in autonomous driving.

行车视频去雾/眩光抑制/低光增强，纯 OpenCV 传统图像处理

Swapping YOLOv5m's CNN backbone for a SwinV2-Tiny transformer to test vehicle detection in rain and snow