ML Anthology

Hyperbolic Image Embeddings

Valentin Khrulkov, Leyla Mirvakhabova, Evgeniya Ustinova, Ivan Oseledets, Victor Lempitsky
CVPR 2020
doi:10.1109/CVPR42600.2020.00645 /cvpr/2020/khrulkov2020cvpr-hyperbolic/

Abstract

Computer vision tasks such as image classification, image retrieval, and few-shot learning are currently dominated by Euclidean and spherical embeddings so that the final decisions about class belongings or the degree of similarity are made using linear hyperplanes, Euclidean distances, or spherical geodesic distances (cosine similarity). In this work, we demonstrate that in many practical scenarios, hyperbolic embeddings provide a better alternative.

PDF CVPR Semantic Scholar

Cite

Text

Khrulkov et al. "Hyperbolic Image Embeddings." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020. doi:10.1109/CVPR42600.2020.00645

Markdown

[Khrulkov et al. "Hyperbolic Image Embeddings." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020.](https://mlanthology.org/cvpr/2020/khrulkov2020cvpr-hyperbolic/) doi:10.1109/CVPR42600.2020.00645

BibTeX

@inproceedings{khrulkov2020cvpr-hyperbolic,
  title     = {{Hyperbolic Image Embeddings}},
  author    = {Khrulkov, Valentin and Mirvakhabova, Leyla and Ustinova, Evgeniya and Oseledets, Ivan and Lempitsky, Victor},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2020},
  doi       = {10.1109/CVPR42600.2020.00645},
  url       = {https://mlanthology.org/cvpr/2020/khrulkov2020cvpr-hyperbolic/}
}