CaltechAUTHORS
  A Caltech Library Service

DiscoBox: Weakly Supervised Instance Segmentation and Semantic Correspondence from Box Supervision

Lan, Shiyi and Yu, Zhiding and Choy, Christopher and Radhakrishnan, Subhashree and Liu, Guilin and Zhu, Yuke and Davis, Larry S. and Anandkumar, Anima (2021) DiscoBox: Weakly Supervised Instance Segmentation and Semantic Correspondence from Box Supervision. . (Unpublished) https://resolver.caltech.edu/CaltechAUTHORS:20210831-203854134

[img] PDF - Submitted Version
See Usage Policy.

2MB

Use this Persistent URL to link to this item: https://resolver.caltech.edu/CaltechAUTHORS:20210831-203854134

Abstract

We introduce DiscoBox, a novel framework that jointly learns instance segmentation and semantic correspondence using bounding box supervision. Specifically, we propose a self-ensembling framework where instance segmentation and semantic correspondence are jointly guided by a structured teacher in addition to the bounding box supervision. The teacher is a structured energy model incorporating a pairwise potential and a cross-image potential to model the pairwise pixel relationships both within and across the boxes. Minimizing the teacher energy simultaneously yields refined object masks and dense correspondences between intra-class objects, which are taken as pseudo-labels to supervise the task network and provide positive/negative correspondence pairs for dense constrastive learning. We show a symbiotic relationship where the two tasks mutually benefit from each other. Our best model achieves 37.9% AP on COCO instance segmentation, surpassing prior weakly supervised methods and is competitive to supervised methods. We also obtain state of the art weakly supervised results on PASCAL VOC12 and PF-PASCAL with real-time inference.


Item Type:Report or Paper (Discussion Paper)
Related URLs:
URLURL TypeDescription
http://arxiv.org/abs/2105.06464arXivDiscussion Paper
ORCID:
AuthorORCID
Liu, Guilin0000-0003-2390-7927
Zhu, Yuke0000-0002-9198-2227
Additional Information:We would like to sincerely thank Xinlong Wang, Zhi Tian, Shuaiyi Huang, Yashar Asgarieh, Jose M. Alvarez, De-An Huang, and other NVIDIA colleagues for the discussion and constructive suggestions.
Record Number:CaltechAUTHORS:20210831-203854134
Persistent URL:https://resolver.caltech.edu/CaltechAUTHORS:20210831-203854134
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:110644
Collection:CaltechAUTHORS
Deposited By: George Porter
Deposited On:01 Sep 2021 14:54
Last Modified:01 Sep 2021 14:54

Repository Staff Only: item control page