Semi-Supervised Semantic Matching

ECCVW 2018 pp. 444-455

doi:10.1007/978-3-030-11015-4_32 /eccvw/2018/laskar2018eccvw-semisupervised/

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-of-the-art on a benchmark semantic matching dataset.

PDF ECCVW Semantic Scholar

Cite

Text

Laskar and Kannala. "Semi-Supervised Semantic Matching." European Conference on Computer Vision Workshops, 2018. doi:10.1007/978-3-030-11015-4_32

Markdown

[Laskar and Kannala. "Semi-Supervised Semantic Matching." European Conference on Computer Vision Workshops, 2018.](https://mlanthology.org/eccvw/2018/laskar2018eccvw-semisupervised/) doi:10.1007/978-3-030-11015-4_32

BibTeX

@inproceedings{laskar2018eccvw-semisupervised,
  title     = {{Semi-Supervised Semantic Matching}},
  author    = {Laskar, Zakaria and Kannala, Juho},
  booktitle = {European Conference on Computer Vision Workshops},
  year      = {2018},
  pages     = {444-455},
  doi       = {10.1007/978-3-030-11015-4_32},
  url       = {https://mlanthology.org/eccvw/2018/laskar2018eccvw-semisupervised/}
}