Color vision deficiency datasets & recoloring evaluation using GANs
- PDF / 10,659,774 Bytes
- 32 Pages / 439.642 x 666.49 pts Page_size
- 54 Downloads / 191 Views
Color vision deficiency datasets & recoloring evaluation using GANs Hongsheng Li1 · Liang Zhang1 · Xiangdong Zhang1 · Meili Zhang1 · Guangming Zhu1 · Peiyi Shen1 · Ping Li2 · Mohammed Bennamoun3 · Syed Afaq Ali Shah3 Received: 23 August 2019 / Revised: 4 June 2020 / Accepted: 6 July 2020 / © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract People with Color Vision Deficiency (CVD) cannot distinguish some color combinations under normal situations. Recoloring becomes a necessary adaptation procedure. In this paper, in order to adaptively find the key color components in an image, we first propose a self-adapting recoloring method with an Improved Octree Quantification Method (IOQM). Second, we design a screening tool of CVD datasets that is used to integrate multiple recoloring methods. Third, a CVD dataset is constructed with the help of our designed screening tool. Our dataset consists of 2313 pairs of training images and 771 pairs of testing images. Fourth, multiple GANs i.e., pix2pix-GAN [1], Cycle-GAN [2], Bicycle-GAN [3] are used for colorblind data conversion. This is the first ever effort in this research area using GANs. Experimental results show that pix2pix-GAN [1] can effectively recolor unrecognizable colors for people with CVD, and we predict that this dataset can provide some help for color blind images recoloring. Datasets and source are available at: https:// github.com/doubletry/pix2pix , https://github.com/doubletry/CycleGAN and https://github. com/doubletry/BicycleGAN. Keywords Color vision deficiency · Recolor · Improved octree quantification method · GAN
1 Introduction With the explosive growth of multimedia technology, perceiving color accurately is becoming a fundamental necessity for effective visual communication. According to the statistics in [51], there are more than 20 million humans with color vision deficiency (CVD) in the world. Obviously, it is necessary to resolve the CVD problem in order for these people to lead a comfortable daily life. Normal vision is based on the response to photons in three different type of cones which are contained in the retina of the eye, and the peak sensitivities Liang Zhang
[email protected]
Extended author information available on the last page of the article.
Multimedia Tools and Applications
of these three types are located in the long wavelength (L), middle wavelength (M), and the short wavelength (S) regions of the visible spectrum, respectively [58]. However, people with CVD are divided from partial or complete deficiency of one or more cones. In fact, there are three major types of CVD: anomalous trichromacy, dichromacy and monochromacy. Anomalous trichromacy results from the shift of the peak sensitivity of the cones, which can be divided into protanomaly, deuteranomaly and tritanomaly, according to the shift spectrum of L−, M−, and S− cones, respectively. Dichromacy is presented when one of the cones is absent, and it can also be divided into protanopia, deuteranopia and tritanopia, respectively. Monoc
Data Loading...
