The Power of Asymmetry in Binary Hashing

TL;DR

This paper demonstrates that using two distinct binary code maps for similarity approximation can produce shorter, more accurate hashes than symmetric mappings, leveraging asymmetry to improve binary hashing performance.

Contribution

It introduces a novel asymmetric hashing approach that improves accuracy and efficiency by employing separate code maps for similarity approximation.

Findings

Asymmetric hashing yields shorter, more accurate binary codes.

Using two distinct code maps improves similarity approximation.

The method outperforms symmetric hashing in experiments.

Abstract

When approximating binary similarity using the hamming distance between short binary hashes, we show that even if the similarity is symmetric, we can have shorter and more accurate hashes by using two distinct code maps. I.e. by approximating the similarity between $x$ and $x'$ as the hamming distance between $f(x)$ and $g(x')$, for two distinct binary codes $f,g$, rather than as the hamming distance between $f(x)$ and $f(x')$.