CVJul 21, 2017

Evaluation of Hashing Methods Performance on Binary Feature Descriptors

arXiv:1707.06825v1
Originality Synthesis-oriented
AI Analysis

This work addresses the problem of efficient binary feature representation for computer vision applications, but it is incremental as it focuses on evaluating existing methods rather than introducing new ones.

The paper evaluated various data-dependent hashing methods to reduce the dimensionality of 512-bit FREAK descriptors, finding performance differences among unsupervised, semi-supervised, and supervised approaches on large labeled datasets.

In this paper we evaluate performance of data-dependent hashing methods on binary data. The goal is to find a hashing method that can effectively produce lower dimensional binary representation of 512-bit FREAK descriptors. A representative sample of recent unsupervised, semi-supervised and supervised hashing methods was experimentally evaluated on large datasets of labelled binary FREAK feature descriptors.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes