Mining Cross-Image Semantics for Weakly Supervised Semantic Segmentation

by   Guolei Sun, et al.

This paper studies the problem of learning semantic segmentation from image-level supervision only. Current popular solutions leverage object localization maps from classifiers as supervision signals, and struggle to make the localization maps capture more complete object content. Rather than previous efforts that primarily focus on intra-image information, we address the value of cross-image semantic relations for comprehensive object pattern mining. To achieve this, two neural co-attentions are incorporated into the classifier to complimentarily capture cross-image semantic similarities and differences. In particular, given a pair of training images, one co-attention enforces the classifier to recognize the common semantics from co-attentive objects, while the other one, called contrastive co-attention, drives the classifier to identify the unshared semantics from the rest, uncommon objects. This helps the classifier discover more object patterns and better ground semantics in image regions. In addition to boosting object pattern learning, the co-attention can leverage context from other related images to improve localization map inference, hence eventually benefiting semantic segmentation learning. More essentially, our algorithm provides a unified framework that handles well different WSSS settings, i.e., learning WSSS with (1) precise image-level supervision only, (2) extra simple single-label data, and (3) extra noisy web data. It sets new state-of-the-arts on all these settings, demonstrating well its efficacy and generalizability. Moreover, our approach ranked 1st place in the Weakly-Supervised Semantic Segmentation Track of CVPR2020 Learning from Imperfect Data Challenge.


page 2

page 6

page 13


Contrastive learning of Class-agnostic Activation Map for Weakly Supervised Object Localization and Semantic Segmentation

While class activation map (CAM) generated by image classification netwo...

Tell Me Where to Look: Guided Attention Inference Network

Weakly supervised learning with only coarse labels can obtain visual exp...

Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi- Supervised Semantic Segmentation

Despite the remarkable progress, weakly supervised segmentation approach...

Regional Semantic Contrast and Aggregation for Weakly Supervised Semantic Segmentation

Learning semantic segmentation from weakly-labeled (e.g., image tags onl...

Frame-to-Frame Aggregation of Active Regions in Web Videos for Weakly Supervised Semantic Segmentation

When a deep neural network is trained on data with only image-level labe...

Unsupervised Part Discovery by Unsupervised Disentanglement

We address the problem of discovering part segmentations of articulated ...

LID 2020: The Learning from Imperfect Data Challenge Results

Learning from imperfect data becomes an issue in many industrial applica...

Please sign up or login with your details

Forgot password? Click here to reset