LATCH: Learned Arrangements of Three Patch Codes

01/15/2015
by   Gil Levi, et al.
0

We present a novel means of describing local image appearances using binary strings. Binary descriptors have drawn increasing interest in recent years due to their speed and low memory footprint. A known shortcoming of these representations is their inferior performance compared to larger, histogram based descriptors such as the SIFT. Our goal is to close this performance gap while maintaining the benefits attributed to binary representations. To this end we propose the Learned Arrangements of Three Patch Codes descriptors, or LATCH. Our key observation is that existing binary descriptors are at an increased risk from noise and local appearance variations. This, as they compare the values of pixel pairs; changes to either of the pixels can easily lead to changes in descriptor values, hence damaging its performance. In order to provide more robustness, we instead propose a novel means of comparing pixel patches. This ostensibly small change, requires a substantial redesign of the descriptors themselves and how they are produced. Our resulting LATCH representation is rigorously compared to state-of-the-art binary descriptors and shown to provide far better performance for similar computation and space requirements.

READ FULL TEXT

page 1

page 8

research
05/13/2023

Illumination-insensitive Binary Descriptor for Visual Measurement Based on Local Inter-patch Invariance

Binary feature descriptors have been widely used in various visual measu...
research
09/13/2016

The CUDA LATCH Binary Descriptor: Because Sometimes Faster Means Better

Accuracy, descriptor size, and the time required for extraction and matc...
research
08/15/2019

Beyond Cartesian Representations for Local Descriptors

The dominant approach for learning local patch descriptors relies on sma...
research
01/20/2015

Constructing Binary Descriptors with a Stochastic Hill Climbing Search

Binary descriptors of image patches provide processing speed advantages ...
research
06/01/2015

Robust Face Recognition with Structural Binary Gradient Patterns

This paper presents a computationally efficient yet powerful binary fram...
research
01/11/2019

Feature Fusion for Robust Patch Matching With Compact Binary Descriptors

This work addresses the problem of learning compact yet discriminative p...
research
07/19/2020

Improving the HardNet Descriptor

In the thesis we consider the problem of local feature descriptor learni...

Please sign up or login with your details

Forgot password? Click here to reset