P-NOC: adversarial training of CAM generating networks for robust weakly supervised semantic segmentation priors
This work addresses the challenge of improving segmentation accuracy in weakly supervised settings, which is incremental as it builds on existing WSSS techniques by combining them in novel ways.
The paper tackled the problem of refining Class Activation Maps (CAMs) for Weakly Supervised Semantic Segmentation (WSSS) by introducing two new techniques, P-NOC and CCAM-H, which involve adversarial training and saliency guidance to produce higher quality pseudo-segmentation masks, achieving competitive results with state-of-the-art methods.
Weakly Supervised Semantic Segmentation (WSSS) techniques explore individual regularization strategies to refine Class Activation Maps (CAMs). In this work, we first analyze complementary WSSS techniques in the literature, their segmentation properties, and the conditions in which they are most effective. Based on these findings, we devise two new techniques: P-NOC and CCAM-H. In the first, we promote the conjoint training of two adversarial CAM generating networks: the generator, which progressively learns to erase regions containing class-specific features, and a discriminator, which is refined to gradually shift its attention to new class discriminant features. In the latter, we employ the high quality pseudo-segmentation priors produced by P-NOC to guide the learning to saliency information in a weakly supervised fashion. Finally, we employ both pseudo-segmentation priors and pseudo-saliency proposals in the random walk procedure, resulting in higher quality pseudo-semantic segmentation masks, and competitive results with the state of the art.