CSFNet: A Cosine Similarity Fusion Network for Real-Time RGB-X Semantic Segmentation of Driving Scenes

Benchmark Model Rank Results
real-time-semantic-segmentation-on-cityscapes-1CSFNet-2#12mIoU: 76.36Frame (fps): 72.3 (3090)
real-time-semantic-segmentation-on-cityscapes-1CSFNet-1#18mIoU: 74.73Frame (fps): 106.1
semantic-segmentation-on-cityscapes-valCSFNet-2#65mIoU: 76.36FPS: 72.3 (3090)
semantic-segmentation-on-cityscapes-valCSFNet-1#71mIoU: 74.73FPS: 106.1 (3090)
semantic-segmentation-on-zju-rgb-pCSFNet-2#5mIoU: 91.40Frame (fps): 75 (3090)
semantic-segmentation-on-zju-rgb-pCSFNet-1#6mIoU: 90.85Frame (fps): 108.5
thermal-image-segmentation-on-mfn-datasetCSFNet-2#4mIOU: 59.98Frame (fps): 72.7 (3090)
thermal-image-segmentation-on-mfn-datasetCSFNet-1#22mIOU: 56.05Frame (fps): 106.3 (3090)