Semantic Segmentation On Densepass

Metrics

mIoU

Results

Performance results of various models on this benchmark

Model Name	mIoU	Paper Title	Repository
PVT (Tiny, FPN)	31.20%	Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions	-
SwiftNet (Merge3)	32.04%	ISSAFE: Improving Semantic Segmentation in Accidents by Fusing Event-based Data	-
CLAN	31.46%	Taking A Closer Look at Domain Shift: Category-level Adversaries for Semantics Consistent Domain Adaptation	-
ERFNet	16.65%	ERFNet: Efficient Residual Factorized ConvNet for Real-time Semantic Segmentation
Seamless (Mapillary)	34.14%	Seamless Scene Segmentation	-
Trans4PASS+ (multi-scale)	57.23%	Behind Every Domain There is a Shift: Adapting Distortion-aware Vision Transformers for Panoramic Semantic Segmentation	-
SIM	44.58%	Differential Treatment for Stuff and Things: A Simple Unsupervised Domain Adaptation Method for Semantic Segmentation	-
SegFormer (MiT-B1)	38.5%	SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers	-
SETR (MLA, Transformer-L)	35.6%	Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers	-
Trans4PASS (single-scale)	55.25%	Bending Reality: Distortion-aware Transformers for Adapting to Panoramic Semantic Segmentation	-
Trans4PASS+ (single-scale)	56.45%	Behind Every Domain There is a Shift: Adapting Distortion-aware Vision Transformers for Panoramic Semantic Segmentation	-
DANet (ResNet-101)	28.5%	Dual Attention Network for Scene Segmentation	-
Fast-SCNN	24.6%	Fast-SCNN: Fast Semantic Segmentation Network	-
ECANet	43.02%	Capturing Omni-Range Context for Omnidirectional Segmentation	-
SegFormer (MiT-B2)	42.4%	SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers	-
USSS (Mapillary)	30.87%	Universal Semi-Supervised Semantic Segmentation	-
USSS (IDD)	26.98%	Universal Semi-Supervised Semantic Segmentation	-
PCS	53.83%	Prototypical Cross-domain Self-supervised Learning for Few-shot Unsupervised Domain Adaptation	-
DNL (ResNet-101)	32.1%	Disentangled Non-Local Neural Networks	-
ASMLP (MiT-B1)	42.05%	AS-MLP: An Axial Shifted MLP Architecture for Vision	-

0 of 36 row(s) selected.