Showing papers by "Charles R. Dyer published in 2009"

Half-integrality based algorithms for cosegmentation of images

Abstract: We study the cosegmentation problem where the objective is to segment the same object (i.e., region) from a pair of images. The segmentation for each image can be cast using a partitioning/segmentation function with an additional constraint that seeks to make the histograms of the segmented regions (based on intensity and texture features) similar. Using Markov random field (MRF) energy terms for the simultaneous segmentation of the images together with histogram consistency requirements using the squared L2 (rather than L1) distance, after linearization and adjustments, yields an optimization model with some interesting combinatorial properties. We discuss these properties which are closely related to certain relaxation strategies recently introduced in computer vision. Finally, we show experimental results of the proposed approach.

A study on automatic age estimation using a large database

Abstract: In this paper we study some problems related to human age estimation using a large database. First, we study the influence of gender on age estimation based on face representations that combine biologically-inspired features with manifold learning techniques. Second, we study age estimation using smaller gender and age groups rather than on all ages. Significant error reductions are observed in both cases. Based on these results, we designed three frameworks for automatic age estimation that exhibit high performance. Unlike previous methods that require manual separation of males and females prior to age estimation, our work is the first to estimate age automatically on a large database. Furthermore, a data fusion approach is proposed using one of the frameworks, which gives an age estimation error more than 40% smaller than previous methods.

Is gender recognition affected by age

Abstract: Gender recognition is important for many applications including human computer interaction (HCI). This paper shows that gender recognition accuracy is affected significantly by the age of the person. Our empirical studies on a large face database of 8,000 images with ages from 0 to 93 years show that gender classification accuracy on adult faces can be 10% higher than that on young or senior faces, evaluated using one of the state-of-the-art methods. We examine aging effects on human faces, which motivates us to investigate which features can incorporate shape and texture variations on faces together with gender encoding. Based on the aging effects, the local binary pattern (LBP) and histograms of oriented gradients (HOG) methods are evaluated for gender characterization with age variation. We also investigate a biologically-inspired method for gender recognition. Overall, no matter what methods are used, the accuracies on adult faces are consistently higher than on young or senior faces. This new finding suggests new efforts in both psychological studies and computational visual recognition for the purpose of HCI applications.

Physically-based Animation Rendering with Markov Chain Monte Carlo

Abstract: Exploring temporal coherence among light transport paths i s very important to remove temporally perception-sensitiv e artifacts in animation rendering. Using the contribution o f a light transport path to all frames in an animation as the sampling distribution function allows us to adapt Markov Ch ain Monte Carlo (MCMC) algorithms to exploit the temporal and spatial coherence among paths in order to generate a p rceptually pleasant animation. A new perturbation technique calledtime perturbationis developed to explore the temporal coherence among paths. Furthermore, in order to make animation rendering plausible, we distribute itera ive computational tasks to a pool of computers for parallel computation. Each task is rendered with a set of parameters a dapted according to the local properties of each task. We demonstrate that this local adaptation does not introduce b ias statistically. The resulting animations are perceptua lly better than those rendered in a frame-by-frame manner.

Estimating correspondence between multiple cameras using joint invariants

Abstract: The joint invariants of the projective group PSL(3,ℝ) on ℝℙ2, the five-point volume cross-ratios, are studied to address the problem of correspondence in a camera network. The distribution of cross-ratios over the unit square as well as in a small local-neighbourhood of a reference point are found to have a heavy tail. No cross ratio value is unique but the collection of five point cross ratios generated by taking all possible combination of five points completely prescribes the curve. Sections of the signature submanifold that admit large enough variation of cross ratios are found to be sufficient in providing correspondence across wide perspectives. Such invariant signatures may be collected independently at cameras with different viewpoints and shared, thereby achieving the registration of objects in the image. Experimental results with license plate database are provided.