Skin color has proven to be a useful and robust cue for face detection, localization and tracking. Image content filtering, content-aware video compression and image color balancing applications can also benefit from automatic detection of skin in images. Numerous techniques for skin color modelling and recognition have been proposed during several past years. A few papers comparing different approaches have been published [Zarit et al. 1999], [Terrillon et al. 2000], [Brand and Mason 2000]. However, a comprehensive survey on the topic is still missing. We try to fill this vacuum by reviewing most widely used methods and techniques and collecting their numerical evaluation results.

A novel approach for real-time skin segmentation in video sequences is described. The approach enables reliable skin segmentation despite wide variation in illumination during tracking.

Abstract—Hyperspectral cameras provide useful discriminants for human face recognition that cannot be obtained by other imaging methods. We examine the utility of using near-infrared hyperspectral images for the recognition of faces over a database of 200 subjects. The hyperspectral images were collected using a CCD camera equipped with a liquid crystal tunable filter to provide 31 bands over the near-infrared (0.7 m-1.0 m). Spectral measurements over the near-infrared allow the sensing of subsurface tissue structure which is significantly different from person to person, but relatively stable over time. The local spectral properties of human tissue are nearly invariant to face orientation and expression which allows hyperspectral discriminants to be used for recognition over a large range of poses and expressions. We describe a face recognition algorithm that exploits spectral measurements for multiple facial tissue types. We demonstrate experimentally that this algorithm can be used to recognize faces over time in the presence of changes in facial pose and expression. Index Terms—Face recognition, hyperspectral. æ

Abstract not found

Statistics-based colour constancy algorithms work well as long as there are many colours in a scene, they fail however when the en- countering scenes comprise few surfaces. In contrast, physics-based algorithms, based on an understanding of physical processes such as highlights and interreflections, are theoretically able to solve for colour constancy even when there are as few as two surfaces in a scene. Un- fortunately, physics-based theories rarely work outside the lab. In this paper we show that a combination of physical and statistical knowledge leads to a surprisingly simple and powerful colour constancy algorithm, one that also works well for images of natural scenes.

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.

In this paper, we propose a probabilistic region classification scheme for natural scene images. In conventional generative methods, a generative model is learnt for each class using all the available training data belonging to that class. However, if an input image has been generated from only a subset of the model support, use of the full model to assign generative probabilities can produce serious artifacts in the probability assignments. This problem arises mainly when the different classes have multimodal distributions with considerable overlap in the feature space. We propose an approach to constrain the class generative probability of a set of newly observed data by exploiting the distribution of the new data itself and using linear weighted mixing. A Kullback -- Leibler Divergence-based fast model selection procedure is also proposed for learning mixture models in a low dimensional feature space. The preliminary results on the natural scene images support the effectiveness of the proposed approach.

The automated detection of humans in computer vision as well as the realistic rendering of people in computer graphics necessitates improved modeling of the human skin color. In this paper we describe the acquisition and modeling of skin reflectance data densely sampled over the entire visible spectrum. The data collected through a spectrograph allows us to explain skin color (and its variations) and to discriminate between human skin and dyes designed to mimic human skin color. We study the approximation of these data using several sets of basis functions. Our study shows that skin reflectance data can best be approximated by a linear combination of Gaussians or their first derivatives. This result has a significant practical impact on optical acquisition devices: the entire visible spectrum of skin reflectance can now be captured with a few filters of optimally chosen central wavelengths and bandwidth.

Statistics-based colour constancy algorithms work well as long as there are many colours in a scene, they fail however when the encountering scenes comprise few surfaces. In contrast, physics-based algorithms, based on an understanding of physical processes such as highlights and interreflections, are theoretically able to solve for colour constancy even when there are as few as two surfaces in a scene. Unfortunately, physics-based theories rarely work outside the lab. In this paper we show that a combination of physical and statistical knowledge leads to a surprisingly simple and powerful colour constancy algorithm, one that also works well for images of natural scenes.

While human skin is relatively easy to detect in controlled environments, detection in uncontrolled settings such as in consumer digital photographs is generally hard. Algorithms need to robustly deal with variations in lighting, color resolution, and imaging noise. This paper proposes a simple generative skin patch model combining shape and color information. The model is parametric and represents the spatial arrangement of skin pixels as compact elliptical regions. Its parameters are estimated by maximizing the mutual information between the model-generated skin pixel distribution and the distribution of skin color as observed in the image. The core of this work is an empirical evaluation on a database of 653 consumer digital photographs. In addition, we investigate the potential of combining our skin detector with state-of-the-art appearance-based face detectors.