GETAM: Gradient-weighted Element-wise Transformer Attention Map for Weakly-supervised Semantic segmentation. (arXiv:2112.02841v2 [cs.CV] UPDATED)

Weakly Supervised Semantic Segmentation (WSSS) is challenging, particularly
when image-level labels are used to supervise pixel level prediction. To bridge
their gap, a Class Activation Map (CAM) is usually generated to provide pixel
level pseudo labels. CAMs in Convolutional Neural Networks suffer from partial
activation ie, only the most discriminative regions are activated. Transformer
based methods, on the other hand, are highly effective at exploring global
context with long range dependency modeling, potentially alleviating the
"partial activation" issue. In this …

arxiv attention cv gradient map segmentation semantic transformer weakly-supervised