Few Shot Semantic Segmentation On Fss 1000 5

Métriques

Mean IoU

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	Mean IoU	Paper Title	Repository
HSNet (HM, ResNet-101)	88.5	HM: Hybrid Masking for Few-Shot Segmentation	-
HSNet (HM, ResNet-50)	88	HM: Hybrid Masking for Few-Shot Segmentation	-
HSNet (ResNet-50)	87.8	Hypercorrelation Squeeze for Few-Shot Segmentation
SegGPT (ViT)	89.3	SegGPT: Segmenting Everything In Context
VAT (ResNet-50)	90.7	Cost Aggregation with 4D Convolutional Swin Transformer for Few-Shot Segmentation
VAT (HM, ResNet-50)	89.9	HM: Hybrid Masking for Few-Shot Segmentation	-
EfficientLab + PRN	-	PatchRefineNet: Improving Binary Segmentation by Incorporating Signals from Optimal Patch-wise Binarization
Annotation-free FSS (Without Annotation,ResNet-50)	86.8	Self-supervised Few-shot Learning for Semantic Segmentation: An Annotation-free Approach
HSNet (VGG-16)	85.8	Hypercorrelation Squeeze for Few-Shot Segmentation
DACM (ResNet-50)	91.6	Doubly Deformable Aggregation of Covariance Matrices for Few-shot Segmentation
DCAMA (Swin-B)	90.4	Dense Cross-Query-and-Support Attention Weighted Mask Aggregation for Few-Shot Segmentation
DACM (ResNet-101)	91.7	Doubly Deformable Aggregation of Covariance Matrices for Few-shot Segmentation
VAT (HM, ResNet-101)	90.5	HM: Hybrid Masking for Few-Shot Segmentation	-
FSS-1000 (VGG-16)	80.12	FSS-1000: A 1000-Class Dataset for Few-Shot Segmentation
GF-SAM	88.9	Bridge the Points: Graph-based Few-shot Segment Anything Semantically
VAT	90.6	Cost Aggregation Is All You Need for Few-Shot Segmentation
HSNet (ResNet-101)	88.5	Hypercorrelation Squeeze for Few-Shot Segmentation
SSP	88.6	Self-Support Few-Shot Semantic Segmentation
VAT (ResNet-101)	90.8	Cost Aggregation with 4D Convolutional Swin Transformer for Few-Shot Segmentation
DCAMA (ResNet-50)	88.8	Dense Cross-Query-and-Support Attention Weighted Mask Aggregation for Few-Shot Segmentation

0 of 22 row(s) selected.