Real Time Semantic Segmentation On Cityscapes

Metriken

Frame (fps)

mIoU

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

Modellname	Frame (fps)	mIoU	Paper Title	Repository
STDC2-75	97.0(1080Ti)	76.8%	Rethinking BiSeNet For Real-time Semantic Segmentation	-
Template-Based-NAS-arch0	19	67.7%	Template-Based Automatic Search of Compact Semantic Segmentation Architectures	-
DeepLab	0.25	63.1%	Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs	-
EDANet	108.7 (1080Ti)	67.3	Efficient Dense Modules of Asymmetric Convolution for Real-Time Semantic Segmentation	-
DDRNet-23-slim	101.6(2080Ti)	77.4%	Deep Dual-resolution Networks for Real-time and Accurate Semantic Segmentation of Road Scenes	-
U-HarDNet-70	-	-	HarDNet: A Low Memory Traffic Network	-
FasterSeg	163.9	71.5%	-	-
ICNet	30.3	70.6%	ICNet for Real-Time Semantic Segmentation on High-Resolution Images	-
PP-LiteSeg-B1	195.3(1080Ti)	73.9%	PP-LiteSeg: A Superior Real-Time Semantic Segmentation Model	-
PP-LiteSeg-B2	102.6(1080Ti)	77.5%	PP-LiteSeg: A Superior Real-Time Semantic Segmentation Model	-
ShelfNet18	59.2	74.8%	ShelfNet for Fast Semantic Segmentation	-
PIDNet-L	31.1(3090)	80.6%	PIDNet: A Real-time Semantic Segmentation Network Inspired by PID Controllers	-
ESNet	63	70.7%	ESNet: An Efficient Symmetric Network for Real-time Semantic Segmentation	-
ENet	76.9	58.3%	ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation	-
ENet + Lovász-Softmax	76.9	63.1%	The Lovász-Softmax loss: A tractable surrogate for the optimization of the intersection-over-union measure in neural networks	-
TD4-BISE18	47.6 (Titan X)	74.9%	Temporally Distributed Networks for Fast Video Semantic Segmentation	-
BiSeNet V2	156	72.6%	BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation	-
RegSeg (no ImageNet pretraining)	30	78.3%	Rethinking Dilated Convolution for Real-time Semantic Segmentation	-
PP-LiteSeg-T1	273.6(1080Ti)	72.0%	PP-LiteSeg: A Superior Real-Time Semantic Segmentation Model	-
HyperSeg-M	36.9	75.8%	HyperSeg: Patch-wise Hypernetwork for Real-time Semantic Segmentation	-

0 of 39 row(s) selected.