Semi-Global Context Network for Semantic Correspondenceopen access
- Authors
- Lee, Ho-Jun; Choi, Hong Tae; Park, Sung Kyu; Park, Ho-Hyun
- Issue Date
- Dec-2021
- Publisher
- IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
- Keywords
- Context fusion; historical averaging; neighborhood consensus network; semantic correspondence; semi-global self-similarity; weakly supervised learning
- Citation
- IEEE ACCESS, v.9, pp 2496 - 2507
- Pages
- 12
- Journal Title
- IEEE ACCESS
- Volume
- 9
- Start Page
- 2496
- End Page
- 2507
- URI
- https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/44013
- DOI
- 10.1109/ACCESS.2020.3046845
- ISSN
- 2169-3536
- Abstract
- Estimating semantic correspondence between pairs of images can be challenging as a result of intra-class variation, background clutter, and repetitive patterns. This paper proposes a convolutional neural network (CNN) that attempts to learn rich semantic representations that contain the global semantic context to enable robust semantic correspondence estimation against intra-class variation and repetitive patterns. We introduce a global context fused feature representation that efficiently employs the global semantic context in estimating semantic correspondence as well as a semi-global self-similarity feature to reduce background clutter-induced distraction in capturing the global semantic context. The proposed network is trained in an end-to-end manner using a weakly supervised loss, which requires a weak level of supervision involving annotation on image pairs. This weakly supervised loss is supplemented with a historical averaging loss to effectively train the network. Our approach decreases running time by a factor of more than four and reduces the training memory requirement by a factor of three and produces competitive or superior results relative to previous approaches on the PF-PASCAL, PF-WILLOW, and TSS benchmarks.
- Files in This Item
-
- Appears in
Collections - College of ICT Engineering > School of Electrical and Electronics Engineering > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.