Few Shot Semantic Segmentation On Coco 20I 5

评估指标

FB-IoU

Mean IoU

评测结果

各个模型在此基准测试上的表现结果

			Paper Title	Repository
SegGPT (ViT)	-	67.9	SegGPT: Segmenting Everything In Context
GF-SAM	-	66.8	Bridge the Points: Graph-based Few-shot Segment Anything Semantically
PGMA-Net (ResNet-101)	79.4	61.8	Visual and Textual Prior Guided Mask Assemble for Few-Shot Segmentation and Beyond	-
DiffewS(SD2.1)	-	60.7	Unleashing the Potential of the Diffusion Model in Few-shot Semantic Segmentation
Matcher(DINOv2)	-	60.7	Matcher: Segment Anything with One Shot Using All-Purpose Feature Matching
FPTrans (DeiT-B/16)	-	58.9	Feature-Proxy Transformer for Few-Shot Segmentation
HMNet (ResNet-50)	77.6	58.9	Hybrid Mamba for Few-Shot Segmentation
DCAMA (Swin-B)	76.9	58.3	Dense Cross-Query-and-Support Attention Weighted Mask Aggregation for Few-Shot Segmentation
DGPNet (ResNet-101)	-	57.9	Dense Gaussian Processes for Few-Shot Segmentation
AENet (ResNet-50)	78.5	57.1	Eliminating Feature Ambiguity for Few-Shot Segmentation
PGMA-Net (ResNet-50)	76.7	57.1	Visual and Textual Prior Guided Mask Assemble for Few-Shot Segmentation and Beyond	-
SCCAN (ResNet-101)	74.8	57	Self-Calibrated Cross Attention Network for Few-Shot Segmentation
MSANet (ResNet-101)	56.8	56.3	MSANet: Multi-Similarity and Attention Guidance for Boosting Few-Shot Segmentation
DGPNet (ResNet-50)	-	56.2	Dense Gaussian Processes for Few-Shot Segmentation
HDMNet (ResNet-50)	77.7	56	Hierarchical Dense Correlation Distillation for Few-Shot Segmentation
MSDNet (ResNet-101)	75.1	55.3	MSDNet: Multi-Scale Decoder for Few-Shot Semantic Segmentation via Transformer-Guided Prototyping
ProtoFormer (ResNet-101)	74.6	54.7	Prototype as Query for Few Shot Semantic Segmentation
MSDNet (ResNet-50)	74.5	54.5	MSDNet: Multi-Scale Decoder for Few-Shot Semantic Segmentation via Transformer-Guided Prototyping
HMNet (VGG-16)	75.5	54.5	Hybrid Mamba for Few-Shot Segmentation
PFENet (SVF, ResNet-50)	-	54.38	Singular Value Fine-tuning: Few-shot Segmentation requires Few-parameters Fine-tuning

0 of 81 row(s) selected.