The Past and the Present of the Color Checker Dataset Misuse

Banić, Nikola; Koščević, Karlo; Subašić, Marko; Lončarić, Sven

doi:10.1109/ispa.2019.8868882

Cited by 4 publications

(2 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The following benchmark datasets have been used to compare the accuracy of the proposed method to the accuracy of other well-known methods: the GreyBall dataset [56], its approximated linear version, eight linear NUS dataset [16], and a newly created dataset, which is presented in more detail in the following subsection. The ColorChecker dataset [25], [57] has not been used to avoid confusion over different results mentioned in numerous publications during ColorChecker's history of various and wrong usage [58], [59], [60], [61]. Since in digital devices illumination estimation is usually performed on linear images [3] i.e.…”

Section: A Experimental Setupmentioning

confidence: 99%

Unsupervised Learning for Color Constancy

Banić

Lončarić

2018

Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Application

View full text Add to dashboard Cite

Most digital camera pipelines use color constancy methods to reduce the influence of illumination and camera sensor on the colors of scene objects. The highest accuracy of color correction is obtained with learning-based color constancy methods, but they require a significant amount of calibrated training images with known ground-truth illumination. Such calibration is time consuming, preferably done for each sensor individually, and therefore a major bottleneck in acquiring high color constancy accuracy. Statistics-based methods do not require calibrated training images, but they are less accurate. In this paper an unsupervised learning-based method is proposed that learns its parameter values after approximating the unknown ground-truth illumination of the training images, thus avoiding calibration. In terms of accuracy the proposed method outperforms all statisticsbased and many learning-based methods. An extension of the method is also proposed, which learns the needed parameters from non-calibrated images taken with one sensor and which can then be successfully applied to images taken with another sensor. This effectively enables inter-camera unsupervised learning for color constancy. Additionally, a new high quality color constancy benchmark dataset with 1707 calibrated images is created, used for testing, and made publicly available. The results are presented and discussed. The source code and the dataset are available at

show abstract

Section: A Experimental Setupmentioning

confidence: 99%

Unsupervised Learning for Color Constancy

Banić

Lončarić

2018

Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Application

View full text Add to dashboard Cite

show abstract

“…Today the CC is used to control the stability of colours in photography, scanning and object acquisition. Nevertheless, different errors occur in colour reproduction, and different works found the CC insufficient to guarantee the robustness of colour reproduction 11,47 . As a consequence, the entire photogrammetric process and object digital reconstruction can be unfaithful.…”

Section: Digital Colour Reproductionmentioning

confidence: 99%

The limits of colorimetry in cultural heritage applications

Plutino

Simone

2020

Coloration Technology

View full text Add to dashboard Cite

Colours in cultural heritage are not only means of expression, but also materials with physical and chemical properties. Since the beginning of the current century, colorimetry has become a fundamental field aiming at restoring, preserving and valorising the art works. There are different colorimetric applications in cultural heritage (eg, from light design to digitisation and modelling), and the variety of cultural material is extremely wide, from statues and frescoes, to old book and films. In this heterogeneous context, the development of instruments, measurements and tools faithful to the original objects are fundamental to correctly reproducing and representing materials, contrast and colours of the works under analysis. In this work, we will analyse the main limits of colorimetry applied in cultural heritage domain considering a variety of applications on physical colour representation and on digital colour reproduction. We will focus on main state‐of‐the‐art studies, regulations and guidelines for objects illumination, colour analysis and digitisation workflow showing some practical example of issues encountered by standard pointwise colorimetry.

show abstract