UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Few-shot Semantic Segmentation with Self-supervision from Pseudo-classes

Li, Y; Data, GWP; Fu, Y; Hu, Y; Prisacariu, VA; (2021) Few-shot Semantic Segmentation with Self-supervision from Pseudo-classes. In: Proceedings of the 32nd British Machine Vision Conference (BMVC) 2021. British Machine Vision Association: Virtual conference. Green open access

[thumbnail of 2110.11742v1.pdf]
Preview
Text
2110.11742v1.pdf - Accepted Version

Download (9MB) | Preview

Abstract

Despite the success of deep learning methods for semantic segmentation, few-shot semantic segmentation remains a challenging task due to the limited training data and the generalisation requirement for unseen classes. While recent progress has been particularly encouraging, we discover that existing methods tend to have poor performance in terms of meanIoU when query images contain other semantic classes besides the target class. To address this issue, we propose a novel self-supervised task that generates random pseudo-classes in the background of the query images, providing extra training data that would otherwise be unavailable when predicting individual target classes. To that end, we adopted superpixel segmentation for generating the pseudo-classes. With this extra supervision, we improved the meanIoU performance of the state-of-the-art method by 2.5% and 5.1% on the one-shot tasks, as well as 6.7% and 4.4% on the five-shot tasks, on the PASCAL-5i and COCO benchmarks, respectively.

Type: Proceedings paper
Title: Few-shot Semantic Segmentation with Self-supervision from Pseudo-classes
Event: 32nd British Machine Vision Conference (BMVC) 2021
Open access status: An open access version is available from UCL Discovery
Publisher version: https://www.bmvc2021-virtualconference.com/assets/...
Language: English
Additional information: This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions.
UCL classification: UCL
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Med Phys and Biomedical Eng
URI: https://discovery.ucl.ac.uk/id/eprint/10137394
Downloads since deposit
5Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item