Zero Shot Learning via Multi-Scale Manifold Regularization

Shay Deutsch, Soheil Kolouri, Kyungnam Kim, Yuri Owechko, Stefano Soatto

CVPR 2017

doi:10.1109/CVPR.2017.562 /cvpr/2017/deutsch2017cvpr-zero/

Abstract

We address zero-shot learning using a new manifold alignment framework based on a localized multi-scale transform on graphs. Our inference approach includes a smoothness criterion for a function mapping nodes on a graph (visual representation) onto a linear space (semantic representation), which we optimize using multi-scale graph wavelets. The robustness of the ensuing scheme allows us to operate with automatically generated semantic annotations, resulting in an algorithm that is entirely free of manual supervision, and yet improves the state-of-the-art as measured on benchmark datasets.

PDF CVPR Semantic Scholar

Cite

Text

Deutsch et al. "Zero Shot Learning via Multi-Scale Manifold Regularization." Conference on Computer Vision and Pattern Recognition, 2017. doi:10.1109/CVPR.2017.562

Markdown

[Deutsch et al. "Zero Shot Learning via Multi-Scale Manifold Regularization." Conference on Computer Vision and Pattern Recognition, 2017.](https://mlanthology.org/cvpr/2017/deutsch2017cvpr-zero/) doi:10.1109/CVPR.2017.562

BibTeX

@inproceedings{deutsch2017cvpr-zero,
  title     = {{Zero Shot Learning via Multi-Scale Manifold Regularization}},
  author    = {Deutsch, Shay and Kolouri, Soheil and Kim, Kyungnam and Owechko, Yuri and Soatto, Stefano},
  booktitle = {Conference on Computer Vision and Pattern Recognition},
  year      = {2017},
  doi       = {10.1109/CVPR.2017.562},
  url       = {https://mlanthology.org/cvpr/2017/deutsch2017cvpr-zero/}
}