Weakly Supervised Semantic Segmentation On 4

Metrics

mIoU

Results

Performance results of various models on this benchmark

Model Name	mIoU	Paper Title	Repository
CoSA (SWIN-B, multi-stage)	53.7	Weakly Supervised Co-training with Swapping Assignments for Semantic Segmentation
CoSA (ViT-B, single-stage)	51.1	Weakly Supervised Co-training with Swapping Assignments for Semantic Segmentation
RS+EPM (ResNet-101, multi-stage)	46.4	RecurSeed and EdgePredictMix: Pseudo-Label Refinement Learning for Weakly Supervised Semantic Segmentation across Single- and Multi-Stage Frameworks
WeakTr (ViT-S, multi-stage)	50.3	WeakTr: Exploring Plain Vision Transformer for Weakly-supervised Semantic Segmentation
PMM(ResNet38, no saliency, no RW)	36.7	Pseudo-mask Matters in Weakly-supervised Semantic Segmentation
MARS (ResNet-101, multi-stage)	49.4	MARS: Model-agnostic Biased Object Removal without Additional Supervision for Weakly-Supervised Semantic Segmentation
L2G (DeepLabV2-ResNet101)	44.2	L2G: A Simple Local-to-Global Knowledge Transfer Framework for Weakly Supervised Semantic Segmentation
FMA-WSSS (Swin-L)	55.4	Foundation Model Assisted Weakly Supervised Semantic Segmentation
FBR	45.6	Fine-grained Background Representation for Weakly Supervised Semantic Segmentation	-
ClusterCAM	41.8	Clustering-Guided Class Activation for Weakly Supervised Semantic Segmentation
URN(ScaleNet-101, no saliency, no RW)	40.8	Uncertainty Estimation via Response Scaling for Pseudo-mask Noise Mitigation in Weakly-supervised Semantic Segmentation
URN(ResNet-101, no saliency, no RW)	40.7	Uncertainty Estimation via Response Scaling for Pseudo-mask Noise Mitigation in Weakly-supervised Semantic Segmentation
EPS	35.7	Railroad is not a Train: Saliency as Pseudo-pixel Supervision for Weakly Supervised Semantic Segmentation
PMM(ScaleNet101, no saliency, no RW)	40.2	Pseudo-mask Matters in Weakly-supervised Semantic Segmentation
URN(Res2Net-101, no saliency, no RW)	41.5	Uncertainty Estimation via Response Scaling for Pseudo-mask Noise Mitigation in Weakly-supervised Semantic Segmentation
ACR-WSSS(DeepLabV2-ResNet101)	45.0	All-pairs Consistency Learning for Weakly Supervised Semantic Segmentation
DSRG	26.0	Weakly-Supervised Semantic Segmentation Network With Deep Seeded Region Growing
SLRNet	35.0	Learning Self-Supervised Low-Rank Network for Single-Stage Weakly and Semi-Supervised Semantic Segmentation
CLIP-ES(DeepLabV2-ResNet101)	45.4	CLIP is Also an Efficient Segmenter: A Text-Driven Approach for Weakly Supervised Semantic Segmentation
OC-CSE(ResNet38, no saliency, no RW)	36.4	Unlocking the Potential of Ordinary Classifier: Class-Specific Adversarial Erasing Framework for Weakly Supervised Semantic Segmentation

0 of 39 row(s) selected.