Domain Adaptation based Object Detection for Autonomous Driving in Foggy and Rainy Weather

Jinlong Li,Runsheng Xu,Xinyu Liu,Jin Ma,Baolu Li,Qin Zou,Jiaqi Ma,Hongkai Yu

2024-08-21

Abstract:Typically, object detection methods for autonomous driving that rely on supervised learning make the assumption of a consistent feature distribution between the training and testing data, this such assumption may fail in different weather conditions. Due to the domain gap, a detection model trained under clear weather may not perform well in foggy and rainy conditions. Overcoming detection bottlenecks in foggy and rainy weather is a real challenge for autonomous vehicles deployed in the wild. To bridge the domain gap and improve the performance of object detection in foggy and rainy weather, this paper presents a novel framework for domain-adaptive object detection. The adaptations at both the image-level and object-level are intended to minimize the differences in image style and object appearance between domains. Furthermore, in order to improve the model's performance on challenging examples, we introduce a novel adversarial gradient reversal layer that conducts adversarial mining on difficult instances in addition to domain adaptation. Additionally, we suggest generating an auxiliary domain through data augmentation to enforce a new domain-level metric regularization. Experimental findings on public benchmark exhibit a substantial enhancement in object detection specifically for foggy and rainy driving scenarios.

Computer Vision and Pattern Recognition

What problem does this paper attempt to address?

The paper aims to address the issue of decreased object detection performance of autonomous vehicles under adverse weather conditions such as fog and rain. Specifically, existing supervised learning-based object detection methods assume that the feature distribution between training data and test data is consistent, but this assumption may not hold under different weather conditions (e.g., foggy and rainy days). Therefore, the paper proposes a new domain adaptation-based object detection framework to bridge the domain gap between different weather conditions and improve detection performance under adverse weather.

The main objectives include:
1. **Reducing the domain gap between different weather conditions**: Minimizing the differences in image style and object appearance under different weather conditions through image-level and object-level adaptive methods.
2. **Introducing Adversarial Gradient Reversal Layer (AdvGRL)**: Conducting adversarial mining to improve model performance on challenging samples.
3. **Generating auxiliary domains**: Utilizing data augmentation techniques to create an auxiliary domain to reinforce domain-level metric regularization, thereby further enhancing the model's robustness and transferability under adverse weather.

Through these methods, the paper aims to significantly improve the object detection performance of autonomous vehicles in foggy and rainy conditions.