Real Time Semantic Segmentation On Camvid

评估指标

mIoU

评测结果

各个模型在此基准测试上的表现结果

		Paper Title
PIDNet-M (Cityscapes-Pretrained)	82.0	PIDNet: A Real-time Semantic Segmentation Network Inspired by PID Controllers
RTFormer-Slim	81.4	RTFormer: Efficient Design for Real-Time Semantic Segmentation with Transformer
RegSeg(Cityscapes-Pretrained)	80.9	Rethinking Dilated Convolution for Real-time Semantic Segmentation
DDRNet-23(Cityscapes-Pretrained)	80.6	Deep Dual-resolution Networks for Real-time and Accurate Semantic Segmentation of Road Scenes
PIDNet-S (Cityscapes-Pretrained)	80.1	PIDNet: A Real-time Semantic Segmentation Network Inspired by PID Controllers
HyperSeg-L	79.1	HyperSeg: Patch-wise Hypernetwork for Real-time Semantic Segmentation
BiSeNet V2-Large(Cityscapes-Pretrained)	78.5	BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation
HyperSeg-S	78.4	HyperSeg: Patch-wise Hypernetwork for Real-time Semantic Segmentation
BiSeNet V2(Cityscapes-Pretrained)	76.7	BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation
TD2-PSP50	76.0	Temporally Distributed Networks for Fast Video Semantic Segmentation
PP-LiteSeg-B	75	PP-LiteSeg: A Superior Real-Time Semantic Segmentation Model
DDRNet-23-slim	74.7	Deep Dual-resolution Networks for Real-time and Accurate Semantic Segmentation of Road Scenes
S^2-FPN34M	74.2	S$^2$-FPN: Scale-ware Strip Attention Guided Feature Pyramid Network for Real-time Semantic Segmentation
Mobile-Seed	73.6%	Mobile-Seed: Joint Semantic Segmentation and Boundary Detection for Mobile Robots
PP-LiteSeg-T	73.3	PP-LiteSeg: A Superior Real-Time Semantic Segmentation Model
BiSeNet V2-Large	73.2	BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation
TD4-PSP18	72.6	Temporally Distributed Networks for Fast Video Semantic Segmentation
BiSeNet V2	72.4	BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation
S^2-FPN34	71.0	S$^2$-FPN: Scale-ware Strip Attention Guided Feature Pyramid Network for Real-time Semantic Segmentation
S^2-FPN18	69.5	S$^2$-FPN: Scale-ware Strip Attention Guided Feature Pyramid Network for Real-time Semantic Segmentation

0 of 28 row(s) selected.