Semantic Segmentation from Image Labels by Reconstruction from Structured Decomposition
This addresses the problem of under-constrained segmentation for computer vision researchers, but appears incremental as it builds on existing WSSS methods.
The paper tackles weakly supervised image segmentation from image tags by framing it as a reconstruction from decomposition using masks, which embeds regularization implicitly, and shows promising results with robustness against background ambiguity.
Weakly supervised image segmentation (WSSS) from image tags remains challenging due to its under-constraint nature. Most mainstream work focus on the extraction of class activation map (CAM) and imposing various additional regularization. Contrary to the mainstream, we propose to frame WSSS as a problem of reconstruction from decomposition of the image using its mask, under which most regularization are embedded implicitly within the framework of the new problem. Our approach has demonstrated promising results on initial experiments, and shown robustness against the problem of background ambiguity. Our code is available at \url{https://github.com/xuanrui-work/WSSSByRec}.