Towards Automatic Skin Tone Classification in Facial Images

2017 ◽  
pp. 299-309
Author(s):  
Diana Borza ◽  
Sergiu Cosmin Nistor ◽  
Adrian Sergiu Darabant
Keyword(s):  
Skin Tone ◽  
Facial Images

scholarly journals Preferred skin tones reproduction of three ethnic groups under different ambient lighting conditions

2020 ◽  
Vol 2020 (28) ◽  
pp. 94-99
Author(s):  
Mingkai Cao ◽  
Ming Ronnier Luo ◽  
Guoxiang Liu
Keyword(s):  
Ethnic Groups ◽  
Large Scale ◽  
Skin Tone ◽  
Chromatic Adaptation ◽  
Lighting Conditions ◽  
Ambient Lighting ◽  
Predicted Probability ◽  
Hue Angle ◽  
Scale Experiment ◽  
Facial Images

A large-scale experiment was conducted to investigate facial image quality on mobile phones. There were 8 original facial images from 4 skin tone types, each included a male and a female image. Each image was captured at 6500K and they were rendered to have 5 CCT (correlated colour temperature) and 5 Duv (the shifts away from the Blackbody locus) levels via CAT02 chromatic adaptation transform to simulate the effect of the images captured under 25 different lighting conditions. Each image was assessed under 9 ambient lighting conditions( including one dark condition) by 90 observers from 3 ethnic groups (Caucasian, Chinese and South Asian), each 30 observers. Preferred facial skin tone ellipse was established by maximizing the correlation coefficient between the model predicted probability and the preference percentage from the visual results. Four types of preferred skin tones had small differences in hue angle and chroma, but concentrated into a small colour region, about [24.7, 46.1°] for Cab* and hab values respectively. All ethnic group preferred images taken under illuminants having high CCT (6500-8000 K). It was also found that the chroma of the preferred skin tones will slightly increase as the ambient lighting CCT decrease.

Skin tone on mobile displays under different ambient lighting

2020 ◽  
Vol 2020 (1) ◽  
pp. 10-14
Author(s):  
Mingkai Cao ◽  
Ming Ronnier Luo
Keyword(s):  
Skin Tone ◽  
Chromatic Adaptation ◽  
Dark Condition ◽  
Mean Values ◽  
Colour Temperature ◽  
Capture Region ◽  
Ambient Lighting ◽  
Temperature Levels ◽  
Facial Images ◽  
Good Agreement

This experiment was aimed to study the preference of mobile phone facial images captured under different simulated ambient lightings. The experiment was carried out by assessing the preference of images for two facial images under 11 lightings (5 correlated colour temperature levels at 2 illuminances plus a dark condition). Forty-five images were processed via CAT02 chromatic adaptation transform to simulate the pictures captured under different light environments. The results revealed that the preferred capture region was between 6500 and 8000K around -0.05 Duv. Furthermore, it was found that the preferred skin tones of all the 45 rendered had good agreement under all the ambient lightings of viewing, i.e.to have mean values of L*, Cab*, hab of [76.3 25.1, 46.4°] units under D65/10 o conditions.

scholarly journals Investigation of effect of skin tone to facial attractiveness

2020 ◽  
Vol 2020 (28) ◽  
pp. 25-29
Author(s):  
Yan Lu ◽  
Jie Yang ◽  
Kaida Xiao ◽  
Michael Pointer ◽  
Changjun Li ◽  
...  
Keyword(s):  
Ethnic Groups ◽  
Ethnic Differences ◽  
Facial Attractiveness ◽  
Skin Tone ◽  
Cultural Effects ◽  
The Aesthetic ◽  
Human Faces ◽  
Different Cultures ◽  
Facial Images

Previous research has shown the perceptual importance of skin tone appearance and how it contributes to perceived facial attractiveness, yet facial-colour perceptions may vary with different ethnic groups. This research was designed to explore the cross-cultural effects of the facial skin tone on perceived attractiveness between Caucasian (CA) and Chinese (CH) observers. 80 images of real human faces were used for facial attractiveness assessment by the two groups of observers using the categorical judgment method. The results showed overall similar preference but fine-scale differences in the perception of their own-ethnic facial images and other-ethnic facial images. Both groups of observers tended to use different criteria when judging the facial tone of different ethnic groups. Our findings show the aesthetic difference of different cultures in perceptions and underline the important role of ethnic differences with respect to skin tone preference.

How processing objectives influence implicit skin tone bias

2005 ◽  
Author(s):  
Larissa Schyrokyj ◽  
Julie Russ ◽  
Keith Maddox
Keyword(s):  
Skin Tone ◽  
Skin Tone Bias

Evaluation of the Human Visual System in Cosmetics Foundation Colour Selection

2020 ◽  
Vol 2020 (1) ◽  
pp. 60-64
Author(s):  
Altynay Kadyrova ◽  
Majid Ansari-Asl ◽  
Eva Maria Valero Benito
Keyword(s):  
Visual System ◽  
Human Visual System ◽  
Skin Tone ◽  
Skin Colour ◽  
Psychophysical Experiment ◽  
Spectral Measurements ◽  
Practical Applications ◽  
Subjective Evaluations ◽  
Colour Measurements ◽  
Colour Quantification

Colour is one of the most important appearance attributes in a variety of fields including both science and industry. The focus of this work is on cosmetics field and specifically on the performance of the human visual system on the selection of foundation makeup colour that best matches with the human skin colour. In many cases, colour evaluations tend to be subjective and vary from person to person thereby producing challenging problems to quantify colour for objective evaluations and measurements. Although many researches have been done on colour quantification in last few decades, to the best of our knowledge, this is the first study to evaluate objectively a consumer's visual system in skin colour matching through a psychophysical experiment under different illuminations exploiting spectral measurements. In this paper, the experiment setup is discussed and the results from the experiment are presented. The correlation between observers' skin colour evaluations by using PANTONE Skin Tone Guide samples and spectroradiometer is assessed. Moreover, inter and intra observer variability are considered and commented. The results reveal differences between nine ethnic groups, between two genders, and between the measurements under two illuminants (i.e.D65 and F (fluorescent)). The results further show that skin colour assessment was done better under D65 than under F illuminant. The human visual system was three times worse than instrument in colour matching in terms of colour difference between skin and PANTONE Skin Tone Guide samples. The observers tend to choose lighter, less reddish, and consequently paler colours as the best match to their skin colour. These results have practical applications. They can be used to design, for example, an application for foundation colour selection based on correlation between colour measurements and human visual system based subjective evaluations.

FACE RECOGNITION USING FUZZY INFERENCE SYSTEM

2009 ◽  
Vol 8 (3) ◽  
pp. 887-897
Author(s):  
Vishal Paika ◽  
Er. Pankaj Bhambri
Keyword(s):  
Neural Networks ◽  
Fuzzy Inference ◽  
Facial Features ◽  
Facial Appearance ◽  
Feature Points ◽  
Inference System ◽  
Detection Techniques ◽  
The Face ◽  
Hair Style ◽  
Facial Images

The face is the feature which distinguishes a person. Facial appearance is vital for human recognition. It has certain features like forehead, skin, eyes, ears, nose, cheeks, mouth, lip, teeth etc which helps us, humans, to recognize a particular face from millions of faces even after a large span of time and despite large changes in their appearance due to ageing, expression, viewing conditions and distractions such as disfigurement of face, scars, beard or hair style. A face is not merely a set of facial features but is rather but is rather something meaningful in its form.In this paper, depending on the various facial features, a system is designed to recognize them. To reveal the outline of the face, eyes, ears, nose, teeth etc different edge detection techniques have been used. These features are extracted in the term of distance between important feature points. The feature set obtained is then normalized and are feed to artificial neural networks so as to train them for reorganization of facial images.

Sign in / Sign up

Export Citation Format

Share Document