. Skin colour detection is an often used cue in human motion tracking, especially in face tracking. Skin colour detection is orientation invariant and fast to process. In this paper skin colour is modelled based on a reectance model of the skin, knowledge about the camera parameters, and the spectrum of the light source. In particular, the location of the skin colour area in the chromaticity plane is estimated for dierent light sources, given known camera characteristics. The model is empirically validated. It has application in adaptive segmentation of skin colour and may have application in the estimation of the colour temperature in camera images containing skin colour.

Colour is an important and useful feature for object tracking and recognition in computer vision. However, it has the difficulty that the colour of the object changes if the illuminant colour changes. But under known illuminant colour it becomes a robust feature. There are more and more computer vision applications tracking humans, for example in interfaces for human computer interaction or automatic camera men, where skin colour is an often used feature. Hence, it would be of significant importance to know the illuminant colour in such applications. This paper proposes a novel method to estimate the current illuminant colour from skin colour observations. The method is based on a physical model of reflections, the assumption that illuminant colours are located close to the Planckian locus, and the knowledge about the camera parameters. The method is empirically tested using real images. The average estimation error of the correlated colour temperature is as small as 180K. Applications are for example in colour based tracking to adapt to changes in lighting and in visualisation to re-render image colours to their appearance under canonical viewing conditions.

The knowledge of the illuminant colour exposing an object or a scenario is very useful for computer vision and allows much more reliable colour based recognition and interpretation. In many computer vision applications humans are important parts of the scenario and the obvious detection of human skin colour is difficult due to varying illumination. Under certain conditions of direct illumination the skin shows highlights. It is well known that highlights of non-homogeneous dielectric materials, such as skin, contain information about the illuminant colour. This paper investigates how accurately the illuminant colour may be estimated from highlights on skin using the Dichromatic Reflection Model. The proposed method extracts automatically the surface vector from a pixel area around a highlight on the tip of a human nose. Experimental results with frontal view images of human faces having different shades of skin taken under different illumination colours and geometries show the estimation of the illuminant colour vector orientation within a few degrees. Keywords: Illuminant Colour, Colour Constancy, Human Skin, Dichromatic Reflection Model 1.