Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Multi-class Token Transformer for Weakly Supervised Semantic Segmentation

About

This paper proposes a new transformer-based framework to learn class-specific object localization maps as pseudo labels for weakly supervised semantic segmentation (WSSS). Inspired by the fact that the attended regions of the one-class token in the standard vision transformer can be leveraged to form a class-agnostic localization map, we investigate if the transformer model can also effectively capture class-specific attention for more discriminative object localization by learning multiple class tokens within the transformer. To this end, we propose a Multi-class Token Transformer, termed as MCTformer, which uses multiple class tokens to learn interactions between the class tokens and the patch tokens. The proposed MCTformer can successfully produce class-discriminative object localization maps from class-to-patch attentions corresponding to different class tokens. We also propose to use a patch-level pairwise affinity, which is extracted from the patch-to-patch transformer attention, to further refine the localization maps. Moreover, the proposed framework is shown to fully complement the Class Activation Mapping (CAM) method, leading to remarkably superior WSSS results on the PASCAL VOC and MS COCO datasets. These results underline the importance of the class token for WSSS.

Lian Xu, Wanli Ouyang, Mohammed Bennamoun, Farid Boussaid, Dan Xu• 2022

Related benchmarks

TaskDatasetResultRank
Semantic segmentationPASCAL VOC 2012 (val)
Mean IoU71.9
2040
Semantic segmentationPASCAL VOC 2012 (test)
mIoU71.6
1342
Change DetectionLEVIR-CD (test)
F1 Score64.55
357
Semantic segmentationPASCAL VOC (val)
mIoU71.9
338
Change DetectionWHU-CD (test)
IoU38.87
286
Semantic segmentationCOCO 2014 (val)
mIoU42
251
Semantic segmentationPascal VOC (test)
mIoU71.6
236
Weakly supervised semantic segmentationPASCAL VOC 2012 (test)
mIoU71.6
158
Weakly supervised semantic segmentationPASCAL VOC 2012 (val)
mIoU71.9
154
Semantic segmentationCOCO (val)
mIoU42
135
Showing 10 of 18 rows

Other info

Follow for update