Solution for CVPR 2024 UG2+ Challenge Track on All Weather Semantic Segmentation
This is an incremental improvement for robust computer vision in challenging weather scenarios, aimed at researchers and practitioners in autonomous driving or surveillance.
The authors tackled semantic segmentation in adverse weather conditions by initializing an InternImage-H backbone with pre-trained weights and enhancing it with Upernet, along with data augmentation, achieving 3rd place in the CVPR 2024 UG2+ Challenge.
In this report, we present our solution for the semantic segmentation in adverse weather, in UG2+ Challenge at CVPR 2024. To achieve robust and accurate segmentation results across various weather conditions, we initialize the InternImage-H backbone with pre-trained weights from the large-scale joint dataset and enhance it with the state-of-the-art Upernet segmentation method. Specifically, we utilize offline and online data augmentation approaches to extend the train set, which helps us to further improve the performance of the segmenter. As a result, our proposed solution demonstrates advanced performance on the test set and achieves 3rd position in this challenge.