Sorry, you need to enable JavaScript to visit this website.

In this paper, we present an automated system for robust biometric recognition based upon sparse representation and dictionary learning. In sparse representation, extracted features from the training data are used to develop a dictionary. Training data of real world applications are likely to be exposed to geometric transformations, which is a big challenge for designing of discriminative dictionaries. Classification is achieved by representing the extracted features of the test data as a linear combination of entries in the dictionary.

The inter prediction decoding is one of the most time consuming modules in modern video decoders, which may significantly limit their real-time capabilities. To circumvent this issue, an efficient acceleration of the HEVC inter prediction decoding module is proposed, by offloading the involved workload to GPU devices. The proposed approach aims at efficiently exploiting the GPU resources by carefully managing the processing within the computational kernels, as well as by optimizing the usage of the complex GPU memory hierarchy.

Removing noise from images usually results in smoothing of edges and areas with discontinuities. Such nonsmooth areas however play a significant role in the perception of image quality. This paper studies the restoration of these regions during denoising. We exploit the fact that the discontinuities in the pixel chromaticity in these regions are less abrupt than those in the pixel luminance. We derive a Bayesian method that estimates the parts of the latent image data in the nonsmooth areas that a denoiser erroneously removes.

A robust, efficient video analytic tool is presented for detection and estimation of driver’s head position and orientation in the SHRP 2 NDS dataset. It is shown that by exploiting the temporal correlations between successive frames in a video, challenging conditions such as lighting variations, and lower resolutions, may be mitigated effectively. The Viola Jones frontal face detector provides initial results. Then, statistical inference methods for spurious faces elimination and missing driver’s head detection are developed.

Over the last few years, industry and academia have been shifting towards computer aided models for seismic interpretation because manual labeling and salt dome delineation is becoming very time consuming and labor intensive as the volume of seismic data is in the order of tera-bytes. In this paper, we have proposed an edge-based active contour with an arc length penalty using level set implementation for interpreter-assisted salt dome segmentation. We have designed an edge function and studied the effect of varying different active contour parameters on salt dome boundary detection.

The most efficient signal edge-preserving smoothing filters, e.g., for denoising, are non-linear. Thus, their acceleration is challenging and is often done in practice by tuning filters parameters, such as increasing the width of the local smoothing neighborhood, resulting in more aggressive smoothing of a single sweep at the cost of increased edge blurring. We propose an alternative technology, accelerating the original filters without tuning, by running them through a conjugate gradient method, not affecting their quality.

Presentation Slides for "Unsupervised Estimation Of Uncertainty For Video Saliency Detection Using Temporal Cues" at GlobalSIP 2015