Revisiting Oxford and Paris: Large-Scale Image Retrieval Benchmarking

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F18%3A00325470" target="_blank" >RIV/68407700:21230/18:00325470 - isvavai.cz</a>

Výsledek na webu

<a href="http://cmp.felk.cvut.cz/~radenfil/publications/Radenovic-CVPR18.pdf" target="_blank" >http://cmp.felk.cvut.cz/~radenfil/publications/Radenovic-CVPR18.pdf</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/CVPR.2018.00598" target="_blank" >10.1109/CVPR.2018.00598</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Revisiting Oxford and Paris: Large-Scale Image Retrieval Benchmarking

Popis výsledku v původním jazyce

In this paper we address issues with image retrieval benchmarking on standard and popular Oxford 5k and Paris 6k datasets. In particular, annotation errors, the size of the dataset, and the level of challenge are addressed: new annotation for both datasets is created with an extra attention to the reliability of the ground truth. Three new protocols of varying difficulty are introduced. The protocols allow fair comparison between different methods, including those using a dataset pre-processing stage. For each dataset, 15 new challenging queries are introduced. Finally, a new set of 1M hard, semi automatically cleaned distractors is selected. An extensive comparison of the state-of-the-art methods is performed on the new benchmark. Different types of methods are evaluated, ranging from local-feature-based to modern CNN based methods. The best results are achieved by taking the best of the two worlds. Most importantly, image retrieval appears far from being solved.

Název v anglickém jazyce

Revisiting Oxford and Paris: Large-Scale Image Retrieval Benchmarking

Popis výsledku anglicky

In this paper we address issues with image retrieval benchmarking on standard and popular Oxford 5k and Paris 6k datasets. In particular, annotation errors, the size of the dataset, and the level of challenge are addressed: new annotation for both datasets is created with an extra attention to the reliability of the ground truth. Three new protocols of varying difficulty are introduced. The protocols allow fair comparison between different methods, including those using a dataset pre-processing stage. For each dataset, 15 new challenging queries are introduced. Finally, a new set of 1M hard, semi automatically cleaned distractors is selected. An extensive comparison of the state-of-the-art methods is performed on the new benchmark. Different types of methods are evaluated, ranging from local-feature-based to modern CNN based methods. The best results are achieved by taking the best of the two worlds. Most importantly, image retrieval appears far from being solved.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

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

Projekt

<a href="/cs/project/LL1303" target="_blank" >LL1303: Vyhledávání vizuálních kategorií ve velkém množství obrázků</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2018

Kód důvěrnosti údajů

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

Název statě ve sborníku

CVPR 2018: Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition

ISBN

978-1-5386-6420-9

ISSN

1063-6919

e-ISSN

2575-7075

Počet stran výsledku

10

Strana od-do

5706-5715

Název nakladatele

IEEE

Místo vydání

Piscataway, NJ

Místo konání akce

Salt Lake City

Datum konání akce

19. 6. 2018

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000457843605089