Incorporating Network Built-in Priors in Weakly-supervised Semantic Segmentation

06/06/2017
by   Fatemeh Sadat Saleh, et al.
0

Pixel-level annotations are expensive and time consuming to obtain. Hence, weak supervision using only image tags could have a significant impact in semantic segmentation. Recently, CNN-based methods have proposed to fine-tune pre-trained networks using image tags. Without additional information, this leads to poor localization accuracy. This problem, however, was alleviated by making use of objectness priors to generate foreground/background masks. Unfortunately these priors either require pixel-level annotations/bounding boxes, or still yield inaccurate object boundaries. Here, we propose a novel method to extract accurate masks from networks pre-trained for the task of object recognition, thus forgoing external objectness modules. We first show how foreground/background masks can be obtained from the activations of higher-level convolutional layers of a network. We then show how to obtain multi-class masks by the fusion of foreground/background ones with information extracted from a weakly-supervised localization network. Our experiments evidence that exploiting these masks in conjunction with a weakly-supervised training loss yields state-of-the-art tag-based weakly-supervised semantic segmentation results.

READ FULL TEXT

page 4

page 5

page 6

page 9

page 10

page 11

page 13

page 14

research
09/02/2016

Built-in Foreground/Background Prior for Weakly-Supervised Semantic Segmentation

Pixel-level annotations are expensive and time consuming to obtain. Henc...
research
08/03/2021

Weakly Supervised Foreground Learning for Weakly Supervised Localization and Detection

Modern deep learning models require large amounts of accurately annotate...
research
08/15/2017

Bringing Background into the Foreground: Making All Classes Equal in Weakly-supervised Video Semantic Segmentation

Pixel-level annotations are expensive and time-consuming to obtain. Henc...
research
04/05/2021

Robust Trust Region for Weakly Supervised Segmentation

Acquisition of training data for the standard semantic segmentation is e...
research
11/24/2016

Full-Resolution Residual Networks for Semantic Segmentation in Street Scenes

Semantic image segmentation is an essential component of modern autonomo...
research
08/21/2020

Robustness and Overfitting Behavior of Implicit Background Models

In this paper, we examine the overfitting behavior of image classificati...
research
10/08/2021

Maximize the Exploration of Congeneric Semantics for Weakly Supervised Semantic Segmentation

With the increase in the number of image data and the lack of correspond...

Please sign up or login with your details

Forgot password? Click here to reset