Background-Aware Pooling and Noise-Aware Loss for Weakly-Supervised Semantic Segmentation

by   Youngmin Oh, et al.

We address the problem of weakly-supervised semantic segmentation (WSSS) using bounding box annotations. Although object bounding boxes are good indicators to segment corresponding objects, they do not specify object boundaries, making it hard to train convolutional neural networks (CNNs) for semantic segmentation. We find that background regions are perceptually consistent in part within an image, and this can be leveraged to discriminate foreground and background regions inside object bounding boxes. To implement this idea, we propose a novel pooling method, dubbed background-aware pooling (BAP), that focuses more on aggregating foreground features inside the bounding boxes using attention maps. This allows to extract high-quality pseudo segmentation labels to train CNNs for semantic segmentation, but the labels still contain noise especially at object boundaries. To address this problem, we also introduce a noise-aware loss (NAL) that makes the networks less susceptible to incorrect labels. Experimental results demonstrate that learning with our pseudo labels already outperforms state-of-the-art weakly- and semi-supervised methods on the PASCAL VOC 2012 dataset, and the NAL further boosts the performance.


page 1

page 7

page 15

page 16

page 17

page 18

page 19


Simpler Does It: Generating Semantic Labels with Objectness Guidance

Existing weakly or semi-supervised semantic segmentation methods utilize...

SpotNet: Self-Attention Multi-Task Network for Object Detection

Humans are very good at directing their visual attention toward relevant...

Weakly Supervised Semantic Image Segmentation with Self-correcting Networks

Building a large image dataset with high-quality object masks for semant...

FGR: Frustum-Aware Geometric Reasoning for Weakly Supervised 3D Vehicle Detection

In this paper, we investigate the problem of weakly supervised 3D vehicl...

Weakly Supervised Semantic Segmentation using Out-of-Distribution Data

Weakly supervised semantic segmentation (WSSS) methods are often built o...

Box-driven Class-wise Region Masking and Filling Rate Guided Loss for Weakly Supervised Semantic Segmentation

Semantic segmentation has achieved huge progress via adopting deep Fully...

Automatic CT Segmentation from Bounding Box Annotations using Convolutional Neural Networks

Accurate segmentation for medical images is important for clinical diagn...