Semantic Segmentation On Camvid

评估指标

Mean IoU

评测结果

各个模型在此基准测试上的表现结果

模型名称	Mean IoU	Paper Title	Repository
RTFormer-Base	82.5	RTFormer: Efficient Design for Real-Time Semantic Segmentation with Transformer
EDANet	66.4	Efficient Dense Modules of Asymmetric Convolution for Real-Time Semantic Segmentation
DenseDecoder	70.9	Dense Decoder Shortcut Connections for Single-Pass Semantic Segmentation	-
LMDNet	63.5	Efficient Road Lane Marking Detection with Deep Learning	-
DSNet-Base	83.32	DSNet: A Novel Way to Use Atrous Convolutions in Semantic Segmentation
ETC-Mobile	76.3	Efficient Semantic Video Segmentation with Per-frame Inference
Template-Based NAS-arch1 (480x360 inputs)	63.2%	Template-Based Automatic Search of Compact Semantic Segmentation Architectures
DeepLabV3Plus + SDCNetAug	81.7%	Improving Semantic Segmentation via Video Propagation and Label Relaxation
DDRNet23	80.6%	Deep Dual-resolution Networks for Real-time and Accurate Semantic Segmentation of Road Scenes
DeepLab-MSc-CRF-LargeFOV	61.6%	Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs
PIDNet-Wider	82.0%	PIDNet: A Real-time Semantic Segmentation Network Inspired by PID Controllers
VideoGCRF	75.2	Deep Spatio-Temporal Random Fields for Efficient Video Segmentation	-
Template-Based NAS-arch0 (480x360 inputs)	63.9%	Template-Based Automatic Search of Compact Semantic Segmentation Architectures
SegNet	46.4%	SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation
DFANet A	64.7%	DFANet: Deep Feature Aggregation for Real-Time Semantic Segmentation
Dilated Convolutions	65.3%	Multi-Scale Context Aggregation by Dilated Convolutions
SIW	83.7	Scaling up Multi-domain Semantic Segmentation with Sentence Embeddings	-
SERNet-Former	84.62	SERNet-Former: Semantic Segmentation by Efficient Residual Network with Attention-Boosting Gates and Attention-Fusion Networks
FC-DenseNet103	66.9%	The One Hundred Layers Tiramisu: Fully Convolutional DenseNets for Semantic Segmentation
BiSeNet	68.7%	BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation

0 of 21 row(s) selected.