Thermal Image Segmentation On Mfn Dataset

Metrics

mIOU

Results

Performance results of various models on this benchmark

Model Name	mIOU	Paper Title	Repository
FRRN	44.2	Full-Resolution Residual Networks for Semantic Segmentation in Street Scenes
UniRGB-IR	59.3	UniRGB-IR: A Unified Framework for RGB-Infrared Semantic Tasks via Adapter Tuning
DANet	41.3	Dual Attention Network for Scene Segmentation
SpiderMesh (ResNet-50)	54.4	SpiderMesh: Spatial-aware Demand-guided Recursive Meshing for RGB-T Semantic Segmentation
CCNet	43.3	CCNet: Criss-Cross Attention for Semantic Segmentation
EGFNet(ConvNeXt)	57.5	EGFNet: Edge-Aware Guidance Fusion Network for RGB–Thermal Urban Scene Parsing
DPLNet	59.3	Efficient Multimodal Semantic Segmentation via Dual-Prompt Learning
CSFNet-2	59.98	CSFNet: A Cosine Similarity Fusion Network for Real-Time RGB-X Semantic Segmentation of Driving Scenes	-
CMX (B4)	59.7	CMX: Cross-Modal Fusion for RGB-X Semantic Segmentation with Transformers
SegNet	42.3	SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation
FEANet	55.3	FEANet: Feature-Enhanced Attention Network for RGB-Thermal Real-time Semantic Segmentation
SegFormer (B2)	53.2	SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers
Sigma-base	61.3	Sigma: Siamese Mamba Network for Multi-Modal Semantic Segmentation
RTFNet	53.2	RTFNet: RGB-Thermal Fusion Network for Semantic Segmentation of Urban Scenes
SegFormer (B4)	54.8	SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers
GEBNet	56.2	GEBNet: Graph-Enhancement Branch Network for RGB-T Scene Parsing	-
EAEFNet (ResNet-152)	58.9	Explicit Attention-Enhanced Fusion for RGB-Thermal Perception Tasks
RSFNet (ResNet-101)	56.2	Residual Spatial Fusion Network for RGB-Thermal Semantic Segmentation	-
CSRPNet	56.0	Channel and Spatial Relation-Propagation Network for RGB-Thermal Semantic Segmentation	-
DooDLeNet	57.3	DooDLeNet: Double DeepLab Enhanced Feature Fusion for Thermal-color Semantic Segmentation	-

0 of 55 row(s) selected.