Weakly Supervised Semantic Segmentation On

Metrics

Mean IoU

Results

Performance results of various models on this benchmark

Model Name	Mean IoU	Paper Title
SIPE (DeepLabV2-ResNet101, no saliency)	68.8	Self-supervised Image-specific Prototype Exploration for Weakly Supervised Semantic Segmentation
PMM(ResNet38)	68.5	Pseudo-mask Matters in Weakly-supervised Semantic Segmentation
LIID (ResNet-101)	66.5	Leveraging Instance-, Image- and Dataset-Level Information for Weakly Supervised Instance Segmentation
URN(ResNet-38, no saliency, no RW)	69.4	Uncertainty Estimation via Response Scaling for Pseudo-mask Noise Mitigation in Weakly-supervised Semantic Segmentation
LIID (Res2Net-101)	69.4	Leveraging Instance-, Image- and Dataset-Level Information for Weakly Supervised Instance Segmentation
AMN (DeepLabV2-ResNet101)	69.5	Threshold Matters in WSSS: Manipulating the Activation for the Robust and Accurate Segmentation Model Against Thresholds
URN(ScaleNet-101, no saliency, no RW)	70.1	Uncertainty Estimation via Response Scaling for Pseudo-mask Noise Mitigation in Weakly-supervised Semantic Segmentation
SEAM+CONTA	66.1	Causal Intervention for Weakly-Supervised Semantic Segmentation
ISIM (ResNet-101)	70.51	ISIM: Iterative Self-Improved Model for Weakly Supervised Segmentation
LIID (ResNet-101, +24K SI)	67.8	Leveraging Instance-, Image- and Dataset-Level Information for Weakly Supervised Instance Segmentation
WeakTr (DeiT-S, multi-stage)	74.0	WeakTr: Exploring Plain Vision Transformer for Weakly-supervised Semantic Segmentation
ResNet-101	66.2	Mining Cross-Image Semantics for Weakly Supervised Semantic Segmentation
SGAN	67.1	Saliency Guided Self-attention Network for Weakly and Semi-supervised Semantic Segmentation
ACR-WSSS(DeepLabV1-ResNet101)	71.2	All-pairs Consistency Learning for Weakly Supervised Semantic Segmentation
Infer-CAM(DeepLabV2-ResNet101)	70.8	Inferring the Class Conditional Response Map for Weakly Supervised Semantic Segmentation
PMM(ResNet38, no saliency, no RW)	68.5	Pseudo-mask Matters in Weakly-supervised Semantic Segmentation
SemPLeS (Swin-L)	83.4	Semantic Prompt Learning for Weakly-Supervised Semantic Segmentation
VWL-L	70.6	Weakly-Supervised Semantic Segmentation with Visual Words Learning and Hybrid Pooling
PMM(Res2Net101, no saliency, no RW)	70.0	Pseudo-mask Matters in Weakly-supervised Semantic Segmentation
OC-CSE(ResNet38,no saliency)	68.4	Unlocking the Potential of Ordinary Classifier: Class-Specific Adversarial Erasing Framework for Weakly Supervised Semantic Segmentation

0 of 73 row(s) selected.