Rethinking CAM in Weakly-Supervised Semantic Segmentation

Rethinking CAM in Weakly-Supervised Semantic Segmentation

Weakly supervised semantic segmentation (WSSS) generally utilizes the Class Activation Map (CAM) to synthesize pseudo-labels. However, the current methods of obtaining CAM focus on salient features of a specific layer, resulting in highlighting the most discriminative regions and further leading to...

Full description

Bibliographic Details
Main Authors: Yuqi Song, Xiaojie Li, Canghong Shi, Shihao Feng, Xin Wang, Yong Luo, Wu Xi
Format: Article
Language:English
Published: IEEE 2022-01-01
Series:IEEE Access
Subjects:
Weakly supervised semantic segmentation
class activation map
ordinary classifier
plug-in
Online Access:https://ieeexplore.ieee.org/document/9941137/

Internet

https://ieeexplore.ieee.org/document/9941137/

Similar Items