P-NOC: adversarial training of CAM generating networks for robust weakly supervised semantic segmentation priors

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: David, Lucas, Pedrini, Helio, Dias, Zanoni
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911793075453952
author David, Lucas
Pedrini, Helio
Dias, Zanoni
author_facet David, Lucas
Pedrini, Helio
Dias, Zanoni
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2305_12522
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle P-NOC: adversarial training of CAM generating networks for robust weakly supervised semantic segmentation priors
David, Lucas
Pedrini, Helio
Dias, Zanoni
Computer Vision and Pattern Recognition
Machine Learning
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.
title P-NOC: adversarial training of CAM generating networks for robust weakly supervised semantic segmentation priors
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2305.12522