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

Class Re-Activation Maps for Weakly-Supervised Semantic Segmentation

About

Extracting class activation maps (CAM) is arguably the most standard step of generating pseudo masks for weakly-supervised semantic segmentation (WSSS). Yet, we find that the crux of the unsatisfactory pseudo masks is the binary cross-entropy loss (BCE) widely used in CAM. Specifically, due to the sum-over-class pooling nature of BCE, each pixel in CAM may be responsive to multiple classes co-occurring in the same receptive field. As a result, given a class, its hot CAM pixels may wrongly invade the area belonging to other classes, or the non-hot ones may be actually a part of the class. To this end, we introduce an embarrassingly simple yet surprisingly effective method: Reactivating the converged CAM with BCE by using softmax cross-entropy loss (SCE), dubbed \textbf{ReCAM}. Given an image, we use CAM to extract the feature pixels of each single class, and use them with the class label to learn another fully-connected layer (after the backbone) with SCE. Once converged, we extract ReCAM in the same way as in CAM. Thanks to the contrastive nature of SCE, the pixel response is disentangled into different classes and hence less mask ambiguity is expected. The evaluation on both PASCAL VOC and MS~COCO shows that ReCAM not only generates high-quality masks, but also supports plug-and-play in any CAM variant with little overhead.

Zhaozheng Chen, Tan Wang, Xiongwei Wu, Xian-Sheng Hua, Hanwang Zhang, Qianru Sun• 2022

Related benchmarks

TaskDatasetResultRank
Semantic segmentationPASCAL VOC 2012 (val)
Mean IoU68.5
2040
Semantic segmentationPASCAL VOC 2012 (test)
mIoU68.4
1342
Semantic segmentationPASCAL VOC (val)
mIoU68.5
338
Semantic segmentationCOCO 2014 (val)
mIoU45
251
Semantic segmentationPascal VOC (test)
mIoU68.4
236
Weakly supervised semantic segmentationPASCAL VOC 2012 (test)
mIoU72.2
158
Weakly supervised semantic segmentationPASCAL VOC 2012 (val)
mIoU71.8
154
Semantic segmentationCOCO (val)
mIoU45
135
CAM seed generationPascal VOC (train)
mIoU54.8
19
Weakly Supervised Class LocalizationPascal VOC (train)
mIoU54.8
15
Showing 10 of 11 rows

Other info

Follow for update