Recurrent transformer networks for semantic correspondence

Seungryong Kim, Stephen Lin, Sangryul Jeon, Dongbo Min, Kwanghoon Sohn

Research output: Contribution to journalConference articlepeer-review

74 Scopus citations

Abstract

We present recurrent transformer networks (RTNs) for obtaining dense correspondences between semantically similar images. Our networks accomplish this through an iterative process of estimating spatial transformations between the input images and using these transformations to generate aligned convolutional activations. By directly estimating the transformations between an image pair, rather than employing spatial transformer networks to independently normalize each individual image, we show that greater accuracy can be achieved. This process is conducted in a recursive manner to refine both the transformation estimates and the feature representations. In addition, a technique is presented for weakly-supervised training of RTNs that is based on a proposed classification loss. With RTNs, state-of-the-art performance is attained on several benchmarks for semantic correspondence.

Original languageEnglish
Pages (from-to)6126-6136
Number of pages11
JournalAdvances in Neural Information Processing Systems
Volume2018-December
StatePublished - 2018
Event32nd Conference on Neural Information Processing Systems, NeurIPS 2018 - Montreal, Canada
Duration: 2 Dec 20188 Dec 2018

Bibliographical note

Publisher Copyright:
© 2018 Curran Associates Inc.All rights reserved.

Fingerprint

Dive into the research topics of 'Recurrent transformer networks for semantic correspondence'. Together they form a unique fingerprint.

Cite this