Object detection in adverse weather conditions for autonomous vehicles using Instruct Pix2Pix
The study uses Instruct Pix2Pix to create weather-based data augmentations for object detection models like Faster R-CNN and YOLOv10. This method generates realistic images under adverse conditions to test and improve the robustness of these systems. Results show that this augmentation significantly enhances model performance in both simulated and real-world driving scenarios.