Learning Binary Codes for High-Dimensional Data Using Bilinear Projections

Yunchao Gong, Sanjiv Kumar, Henry A. Rowley, Svetlana Lazebnik

CVPR 2013

doi:10.1109/CVPR.2013.69 /cvpr/2013/gong2013cvpr-learning/

Abstract

Recent advances in visual recognition indicate that to achieve good retrieval and classification accuracy on largescale datasets like ImageNet, extremely high-dimensional visual descriptors, e.g., Fisher Vectors, are needed. We present a novel method for converting such descriptors to compact similarity-preserving binary codes that exploits their natural matrix structure to reduce their dimensionality using compact bilinear projections instead of a single large projection matrix. This method achieves comparable retrieval and classification accuracy to the original descriptors and to the state-of-the-art Product Quantization approach while having orders of magnitude faster code generation time and smaller memory footprint.

PDF CVPR Semantic Scholar

Cite

Text

Gong et al. "Learning Binary Codes for High-Dimensional Data Using Bilinear Projections." Conference on Computer Vision and Pattern Recognition, 2013. doi:10.1109/CVPR.2013.69

Markdown

[Gong et al. "Learning Binary Codes for High-Dimensional Data Using Bilinear Projections." Conference on Computer Vision and Pattern Recognition, 2013.](https://mlanthology.org/cvpr/2013/gong2013cvpr-learning/) doi:10.1109/CVPR.2013.69

BibTeX

@inproceedings{gong2013cvpr-learning,
  title     = {{Learning Binary Codes for High-Dimensional Data Using Bilinear Projections}},
  author    = {Gong, Yunchao and Kumar, Sanjiv and Rowley, Henry A. and Lazebnik, Svetlana},
  booktitle = {Conference on Computer Vision and Pattern Recognition},
  year      = {2013},
  doi       = {10.1109/CVPR.2013.69},
  url       = {https://mlanthology.org/cvpr/2013/gong2013cvpr-learning/}
}