IRUS Total

HPatches: A benchmark and evaluation of handcrafted and learned local descriptors

File Description SizeFormat 
hpatches19.pdfAccepted version2.02 MBAdobe PDFView/Open
Title: HPatches: A benchmark and evaluation of handcrafted and learned local descriptors
Authors: Balntas, V
Lenc, K
Vedaldi, A
Tuytelaars, T
Matas, J
Mikolajczyk, K
Item Type: Journal Article
Abstract: In this paper, a novel benchmark is introduced for evaluating local image descriptors. We demonstrate limitations of the commonly used datasets and evaluation protocols, that lead to ambiguities and contradictory results in the literature. Furthermore, these benchmarks are nearly saturated due to the recent improvements in local descriptors obtained by learning from large annotated datasets. To address these issues, we introduce a new large dataset suitable for training and testing modern descriptors, together with strictly defined evaluation protocols in several tasks such as matching, retrieval and verification. This allows for more realistic, thus more reliable comparisons in different application scenarios. We evaluate the performance of several state-of-the-art descriptors and analyse their properties. We show that a simple normalisation of traditional hand-crafted descriptors is able to boost their performance to the level of deep learning based descriptors once realistic benchmarks are considered. Additionally we specify a protocol for learning and evaluating using cross validation. We show that when training state-of-the-art descriptors on this dataset, the traditional verification task is almost entirely saturated.
Issue Date: 10-May-2019
Date of Acceptance: 1-May-2019
URI: http://hdl.handle.net/10044/1/77898
DOI: 10.1109/tpami.2019.2915233
ISSN: 0162-8828
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Start Page: 1
End Page: 1
Journal / Book Title: IEEE Transactions on Pattern Analysis and Machine Intelligence
Copyright Statement: © 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Sponsor/Funder: Engineering & Physical Science Research Council (EPSRC)
Engineering & Physical Science Research Council (EPSRC)
Funder's Grant Number: EP/K01904X/2
Keywords: 0801 Artificial Intelligence and Image Processing
0806 Information Systems
0906 Electrical and Electronic Engineering
Artificial Intelligence & Image Processing
Publication Status: Published online
Online Publication Date: 2019-05-10
Appears in Collections:Electrical and Electronic Engineering

Unless otherwise indicated, items in Spiral are protected by copyright and are licensed under a Creative Commons Attribution NonCommercial NoDerivatives License.

Creative Commons