English

Fast Binary Embedding via Circulant Downsampled Matrix -- A Data-Independent Approach

Information Theory 2016-01-26 v1 Computer Vision and Pattern Recognition Machine Learning math.IT

Abstract

Binary embedding of high-dimensional data aims to produce low-dimensional binary codes while preserving discriminative power. State-of-the-art methods often suffer from high computation and storage costs. We present a simple and fast embedding scheme by first downsampling N-dimensional data into M-dimensional data and then multiplying the data with an MxM circulant matrix. Our method requires O(N +M log M) computation and O(N) storage costs. We prove if data have sparsity, our scheme can achieve similarity-preserving well. Experiments further demonstrate that though our method is cost-effective and fast, it still achieves comparable performance in image applications.

Keywords

Cite

@article{arxiv.1601.06342,
  title  = {Fast Binary Embedding via Circulant Downsampled Matrix -- A Data-Independent Approach},
  author = {Sung-Hsien Hsieh and Chun-Shien Lu and Soo-Chang Pei},
  journal= {arXiv preprint arXiv:1601.06342},
  year   = {2016}
}

Comments

8 pages, 4 figures, 4 tables

R2 v1 2026-06-22T12:35:31.603Z