Editing Affective computing (section)

====Facial color====

=====Overview=====

The surface of the human face is innervated with a large network of blood vessels. Blood flow variations in these vessels yield visible color changes on the face. Whether or not facial emotions activate facial muscles, variations in blood flow, blood pressure, glucose levels, and other changes occur. Also, the facial color signal is independent from that provided by facial muscle movements.<ref name="face">{{cite journal | last1=Benitez-Quiroz | first1=Carlos F. | last2=Srinivasan | first2=Ramprakash | last3=Martinez | first3=Aleix M. | title=Facial color is an efficient mechanism to visually transmit emotion | journal=Proceedings of the National Academy of Sciences | volume=115 | issue=14 | date=2018-03-19 | doi=10.1073/pnas.1716084115 | pages=3581–3586| pmid=29555780 | pmc=5889636 | bibcode=2018PNAS..115.3581B | doi-access=free }}</ref>

=====Methodology=====

Approaches are based on facial color changes. Delaunay triangulation is used to create the triangular local areas. Some of these triangles which define the interior of the mouth and eyes (sclera and iris) are removed. Use the left triangular areas’ pixels to create feature vectors.<ref name="face"/> It shows that converting the pixel color of the standard RGB color space to a color space such as oRGB color space<ref name="orgb">{{cite journal | last1=Bratkova | first1=Margarita | last2=Boulos | first2=Solomon | last3=Shirley | first3=Peter | title=oRGB: A Practical Opponent Color Space for Computer Graphics | journal=IEEE Computer Graphics and Applications | volume=29 | issue=1 | year=2009 | doi=10.1109/mcg.2009.13 | pages=42–55| pmid=19363957 | s2cid=16690341 }}</ref> or LMS channels perform better when dealing with faces.<ref name="mec">Hadas Shahar, [[Hagit Hel-Or]], [http://openaccess.thecvf.com/content_ICCVW_2019/papers/CVPM/Shahar_Micro_Expression_Classification_using_Facial_Color_and_Deep_Learning_Methods_ICCVW_2019_paper.pdf Micro Expression Classification using Facial Color and Deep Learning Methods], The IEEE International Conference on Computer Vision (ICCV), 2019, pp. 0–0.</ref> So, map the above vector onto the better color space and decompose into red-green and yellow-blue channels. Then use deep learning methods to find equivalent emotions.