CV 3315 Is All You Need : Semantic Segmentation Competition
This work addresses the challenge of semantic segmentation for urban environments using vehicle camera views, but it is incremental as it primarily applies and benchmarks existing transformer methods without introducing new techniques.
The paper tackled the problem of semantic segmentation in urban scenes with highly imbalanced class distributions by evaluating transformer-based methods, particularly SegFormer, to balance performance and efficiency, achieving up to 80.2% mIoU with SegFormer-B5 and selecting SegFormer-B2 as the competition candidate with 78.5% mIoU and 50.6 GFLOPS.
This competition focus on Urban-Sense Segmentation based on the vehicle camera view. Class highly unbalanced Urban-Sense images dataset challenge the existing solutions and further studies. Deep Conventional neural network-based semantic segmentation methods such as encoder-decoder architecture and multi-scale and pyramid-based approaches become flexible solutions applicable to real-world applications. In this competition, we mainly review the literature and conduct experiments on transformer-driven methods especially SegFormer, to achieve an optimal trade-off between performance and efficiency. For example, SegFormer-B0 achieved 74.6% mIoU with the smallest FLOPS, 15.6G, and the largest model, SegFormer- B5 archived 80.2% mIoU. According to multiple factors, including individual case failure analysis, individual class performance, training pressure and efficiency estimation, the final candidate model for the competition is SegFormer- B2 with 50.6 GFLOPS and 78.5% mIoU evaluated on the testing set. Checkout our code implementation at https://vmv.re/cv3315.