Weakly Supervised Instance Segmentation Based on Two-Stage Transfer Learning
Open Access
- 31 January 2020
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 24135-24144
Abstract
Weakly supervised instance segmentation, which could greatly decrease financial and time cost, is one of fundamental computer vision tasks. State-of-the-art methods mainly concentrate on improving the quality of generated pixel level labels, namely masks, using complex traditional segmentation methods, and ignore the effect of the quality of generated masks. Namely, the masks of small object instances tend to be invalid, which would degrades the performance of instance segmentation. In this paper, we propose a two-stage transfer learning framework for weakly supervised instance segmentation. We explicitly discriminate the invalid and valid generated masks, and just utilize the valid masks for training to avoid the interference of invalid ones. We use a network-based transfer learning strategy to effectively utilize all useful information, including category labels and bounding-box information of all objects and valid generated masks. Besides, we further use a feature-mapping-based transfer learning strategy to improve the performance of small object instance segmentation. We demonstrate the effectiveness of the proposed method on the PASCAL VOC 2012, and the experimental results show that our proposed method is effective and outperforms state-of-the-art methods.Keywords
Funding Information
- National Natural Science Foundation of China (61571071, 61906025)
- Chongqing Research Program of Basic Research and Frontier Technology (cstc2018jcyjAX0227)
- Chongqing Municipal Education Commission (KJQN201900607)
- Chongqing University of Posts and Telecommunications (xxhyf2019-01)
- Chongqing University of Posts and Telecommunications
This publication has 49 references indexed in Scilit:
- Learning to segment under various forms of weak supervision2015
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- Person-Specific Face Antispoofing With Subject Domain AdaptationIEEE Transactions on Information Forensics and Security, 2015
- Learning and Transferring Mid-level Image Representations Using Convolutional Neural Networks2014
- Transfer learning for activity recognition: a surveyKnowledge and Information Systems, 2013
- Cross-language knowledge transfer using multilingual deep neural network with shared hidden layers2013
- Cross-domain activity recognition via transfer learningPervasive and Mobile Computing, 2011
- The Pascal Visual Object Classes (VOC) ChallengeInternational Journal of Computer Vision, 2009
- "GrabCut"ACM Transactions on Graphics, 2004
- Improve Multi-Instance Neural Networks through Feature SelectionNeural Processing Letters, 2004