Sandra Skaff

Learning moods and emotions from color combinations

In this paper, we tackle the problem of associating combinations of colors to abstract categories (e.g. capricious, classic, cool, delicate, etc.). It is evident that such concepts would be difficult to distinguish using single colors, therefore we consider combinations of colors or color palettes. We leverage two novel databases for color palettes and we learn categorization models using low and high level descriptors. Preliminary results show that Fisher representation based on GMMs is the most rewarding strategy in terms of classification performance over a baseline model. We also suggest a process for cleaning weakly annotated data, whilst preserving the visual coherence of categories. Finally, we demonstrate how learning abstract categories on color palettes can be used in the application of color transfer, personalization and image re-ranking.

Anomaly Detection for Video Surveillance Applications

We investigate the problem of anomaly detection for video surveillance applications. In our approach, we use a compression-based similarity measure to determine similarity between images in a video sequence. Images that are sufficiently dissimilar are deemed anomalous and stored to be compared against subsequent images in the sequence. The goal of our research is two-fold; in addition to detecting anomalous images, the issue of heavy computational and storage resource demands is addressed

Active Bayesian color constancy with non-uniform sensors

The human impression of the color of an object is the same when viewed foveally or peripherally, despite the non-uniformity of the spatial distribution of photoreceptors on the retina. We propose that this nonuniformity can be used to attain color constancy, the perception of a constant surface color under varying illumination. We develop a multi-sensor Bayesian technique that solves for the surface reflectance and lighting parameters of a bilinear model by sequentially acquiring measurements from independent sensors. We present two cases: (i) two sets of sensors, each with different spectral sensitivities, (ii) a continuous variation in the spectral sensitivities across the sensor array.

A spectral theory of color perception

The paper adopts the philosophical stance that colors are real and can be identified with spectral models based on the photoreceptor signals. A statistical setting represents spectral profiles as probability density functions. This permits the use of analytic tools from the field of information geometry to determine a new kind of color space and structure deriving therefrom. In particular, the metric of the color space is shown to be the Fisher information matrix. A maximum entropy technique for spectral modeling is proposed that takes into account measurement noise. Theoretical predictions provided by our approach are compared with empirical colorfulness and color similarity data.

A sequential Bayesian approach to color constancy using non-uniform filters

This paper introduces a non-uniform filter formulation into the Brainard and Freeman Bayesian color constancy technique. The formulation comprises sensor measurements taken through a non-uniform filter, of spatially-varying spectral sensitivity, placed on the camera lens. The main goal of this paper is twofold. First, it presents a framework in which sensor measurements obtained through a non-uniform filter can be sequentially incorporated into the Bayesian probabilistic formulation. Second, it shows that such additional measurements obtained reduce the effect of the prior in Bayesian color constancy. For the purposes of testing the proposed framework, we use a filter formulation of two portions of different spectral sensitivities. We show through experiments on real data that improvement in the parameter estimation can be obtained inexpensively by sequentially incorporating additional information obtained from the sensor through the different portions of a filter by Bayesian chaining. We also show that our approach outperforms previous approaches in the literature.

Estimating Surface Reflectance Spectra for Underwater Color Vision

This paper introduces a novel mathematical approach to surface spectral reflectance estimation in unknown underwater environments using uncalibrated color cameras. The approach derives surface spectral estimates without explicitly modeling the underwater medium characteristics such as light scattering and absorption. The latter two phenomena are dependent upon two parameters, which are the distance of the object from the camera and the depth of the object in water. The proposed approach does not require these parameters to be specified in advance. Spectral models are useful for underwater applications, where subtle differences in color need to be distinguished. Such models are also useful for fusing information from multiple images. We show that the proposed approach yields promising results.