Sparse Output Coding for Large-Scale Visual Recognition

CVPR 2013

doi:10.1109/CVPR.2013.430 /cvpr/2013/zhao2013cvpr-sparse/

Abstract

Many vision tasks require a multi-class classifier to discriminate multiple categories, on the order of hundreds or thousands. In this paper, we propose sparse output coding, a principled way for large-scale multi-class classification, by turning high-cardinality multi-class categorization into a bit-by-bit decoding problem. Specifically, sparse output coding is composed of two steps: efficient coding matrix learning with scalability to thousands of classes, and probabilistic decoding. Empirical results on object recognition and scene classification demonstrate the effectiveness of our proposed approach.

PDF CVPR Semantic Scholar

Cite

Text

Zhao and Xing. "Sparse Output Coding for Large-Scale Visual Recognition." Conference on Computer Vision and Pattern Recognition, 2013. doi:10.1109/CVPR.2013.430

Markdown

[Zhao and Xing. "Sparse Output Coding for Large-Scale Visual Recognition." Conference on Computer Vision and Pattern Recognition, 2013.](https://mlanthology.org/cvpr/2013/zhao2013cvpr-sparse/) doi:10.1109/CVPR.2013.430

BibTeX

@inproceedings{zhao2013cvpr-sparse,
  title     = {{Sparse Output Coding for Large-Scale Visual Recognition}},
  author    = {Zhao, Bin and Xing, Eric P.},
  booktitle = {Conference on Computer Vision and Pattern Recognition},
  year      = {2013},
  doi       = {10.1109/CVPR.2013.430},
  url       = {https://mlanthology.org/cvpr/2013/zhao2013cvpr-sparse/}
}