Universal Weakly Supervised Segmentation by Pixel-to-Segment Contrastive LearningDownload PDF

Published: 12 Jan 2021, Last Modified: 03 Apr 2024ICLR 2021 PosterReaders: Everyone
Keywords: weakly supervised representation learning, representation learning for computer vision, metric learning, semantic segmentation
Abstract: Weakly supervised segmentation requires assigning a label to every pixel based on training instances with partial annotations such as image-level tags, object bounding boxes, labeled points and scribbles. This task is challenging, as coarse annotations (tags, boxes) lack precise pixel localization whereas sparse annotations (points, scribbles) lack broad region coverage. Existing methods tackle these two types of weak supervision differently: Class activation maps are used to localize coarse labels and iteratively refine the segmentation model, whereas conditional random fields are used to propagate sparse labels to the entire image. We formulate weakly supervised segmentation as a semi-supervised metric learning problem, where pixels of the same (different) semantics need to be mapped to the same (distinctive) features. We propose 4 types of contrastive relationships between pixels and segments in the feature space, capturing low-level image similarity, semantic annotation, co-occurrence, and feature affinity They act as priors; the pixel-wise feature can be learned from training images with any partial annotations in a data-driven fashion. In particular, unlabeled pixels in training images participate not only in data-driven grouping within each image, but also in discriminative feature learning within and across images. We deliver a universal weakly supervised segmenter with significant gains on Pascal VOC and DensePose. Our code is publicly available at https://github.com/twke18/SPML.
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics
One-sentence Summary: We propose a unified pixel-to-segment contrastive learning loss formulation for weakly supervised semantic segmentation with various types of annotations.
Code: [![github](/images/github_icon.svg) twke18/SPML](https://github.com/twke18/SPML)
Data: [DensePose](https://paperswithcode.com/dataset/densepose), [PASCAL VOC 2012 test](https://paperswithcode.com/dataset/pascal-voc-2012-test), [VOC 2012](https://paperswithcode.com/dataset/voc-2012)
13 Replies