The ColorChecker dataset is one of the most widely used image sets for evaluating and ranking illuminant estimation algorithms. However, this single set of images has at least 3 different sets of ground-truth (i.e. correct answers) associated with it. In the literature it is often asserted that one algorithm is better than another when the algorithms in question have been tuned and tested with the different ground-truths. In this short correspondence we present some of the background as to why the 3 existing ground-truths are different and go on to make a new single and recommended set of correct answers. Experiments reinforce the importance of this work in that we show that the total ordering of a set of algorithms may be reversed depending on whether we use the new or legacy ground-truth data.

Hemrit, G., Finlayson, G., Gijsenij, A., Gehler, P., Bianco, S., Drew, M., et al. (2020). Providing a Single Ground-truth for Illuminant Estimation for the ColorChecker Dataset. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 42(5 (1 May 2020)), 1286-1287 [10.1109/TPAMI.2019.2919824].

Providing a Single Ground-truth for Illuminant Estimation for the ColorChecker Dataset

Bianco, Simone;
2020

Abstract

The ColorChecker dataset is one of the most widely used image sets for evaluating and ranking illuminant estimation algorithms. However, this single set of images has at least 3 different sets of ground-truth (i.e. correct answers) associated with it. In the literature it is often asserted that one algorithm is better than another when the algorithms in question have been tuned and tested with the different ground-truths. In this short correspondence we present some of the background as to why the 3 existing ground-truths are different and go on to make a new single and recommended set of correct answers. Experiments reinforce the importance of this work in that we show that the total ordering of a set of algorithms may be reversed depending on whether we use the new or legacy ground-truth data.
Articolo in rivista - Articolo scientifico
Color Constancy,Illuminant Estimation,Algorithms Evaluation
English
1-lug-2019
2020
42
5 (1 May 2020)
1286
1287
8751985
open
Hemrit, G., Finlayson, G., Gijsenij, A., Gehler, P., Bianco, S., Drew, M., et al. (2020). Providing a Single Ground-truth for Illuminant Estimation for the ColorChecker Dataset. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 42(5 (1 May 2020)), 1286-1287 [10.1109/TPAMI.2019.2919824].
File in questo prodotto:
File Dimensione Formato  
10281-255406.pdf

accesso aperto

Tipologia di allegato: Publisher’s Version (Version of Record, VoR)
Dimensione 415.48 kB
Formato Adobe PDF
415.48 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/255406
Citazioni
  • Scopus 14
  • ???jsp.display-item.citation.isi??? 10
Social impact