Semi-Supervised Semantic Matching.

Zakaria Laskar, Juho Kannala; Proceedings of the European Conference on Computer Vision (ECCV) Workshops, 2018, pp. 0-0

Abstract


Convolutional neural networks (CNNs) have been successfully applied to solve the problem of correspondence estimation between semantically related images. Due to non-availability of large training datasets, existing methods resort to self-supervised or unsupervised training paradigm. In this paper we propose a semi-supervised learning framework that imposes cyclic consistency constraint on unlabeled image pairs. Together with the supervised loss the proposed model achieves state-ofthe-art on a benchmark semantic matching dataset.

Related Material


[pdf] [arXiv]
[bibtex]
@InProceedings{Laskar_2018_ECCV_Workshops,
author = {Laskar, Zakaria and Kannala, Juho},
title = {Semi-Supervised Semantic Matching.},
booktitle = {Proceedings of the European Conference on Computer Vision (ECCV) Workshops},
month = {September},
year = {2018}
}