Working hard to know your neighbor's margins: Local descriptor learning loss

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F17%3A00317310" target="_blank" >RIV/68407700:21230/17:00317310 - isvavai.cz</a>

Result on the web

<a href="http://papers.nips.cc/paper/7068-working-hard-to-know-your-neighbors-margins-local-descriptor-learning-loss" target="_blank" >http://papers.nips.cc/paper/7068-working-hard-to-know-your-neighbors-margins-local-descriptor-learning-loss</a>

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Working hard to know your neighbor's margins: Local descriptor learning loss

Original language description

We introduce a loss for metric learning, which is inspired by the Lowe's matching criterion for SIFT. We show that the proposed loss, that maximizes the distance between the closest positive and closest negative example in the batch, is better than complex regularization methods; it works well for both shallow and deep convolution network architectures. Applying the novel loss to the L2Net CNN architecture results in a compact descriptor named HardNet. It has the same dimensionality as SIFT (128) and shows state-of-art performance in wide baseline stereo, patch verification and instance retrieval benchmarks.

Czech name

—

Czech description

—

Type

D - Article in proceedings

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

Result was created during the realization of more than one project. More information in the Projects tab.

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Publication year

2017

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection

Advances in Neural Information Processing Systems 30

ISBN

—

ISSN

1049-5258

e-ISSN

1049-5258

Number of pages

12

Pages from-to

4827-4838

Publisher name

Neural Information Processing Systems (NIPS) Foundation

Place of publication

—

Event location

LONG BEACH

Event date

Dec 4, 2017

Type of event by nationality

WRD - Celosvětová akce

UT code for WoS article

000452649404087