Standard Gaussian mixture modelling does not possess time sequence information (TSI) other than that which might be embedded in the acoustic features. Dynamic time warping relates directly to TSI, time-warping two sequences of features into alignment. Here, a hybrid system embedding DTW into a GMM is presented. Improved automatic speaker verification performance is demonstrated. Testing 1000 sp...

This paper describes a comparison of approaches for time series classification. Our comparisons included two different outlier removal methods (discords and reverse nearest neighbor), two different distance measures (Euclidean distance and dynamic time warping), and two different classification algorithms (k nearest neighbor and support vector machines). An algorithm for semi-supervised learnin...

In this paper we present results that indicate that the formant frequencies between di erent speakers scale di erently at di erent frequencies. Based on our experiments on speech data, we then numerically compute a universal frequencywarping function, to make the scale-factor independent of frequency in the warped domain. The proposed warping function is found to be similar to the mel-scale, wh...

The Peirce quincuncial projection is a mapping of the surface of a sphere to the interior of a square. It is a conformal mapping except for four points on the equator. These points of non-conformality cause significant artifacts in photographic applications. In this paper, we propose an algorithm and userinterface to mitigate these artifacts. Moreover, in order to facilitate an interactive user...

Dynamic Time Warping (DTW) is a time series distance measure that allows non-linear alignments between series. Constraints on the in form of windows and weights have been introduced because unconstrained DTW too permissive its alignments. However, windowing introduces crude step function, allowing flexibility within window, none beyond it. While not entailing multiplicative weight relative to d...

We are investigating methods to accelerate rendering of architectural walkthroughs. In this paper, we improve upon a cells and portals framework by using images to replace geometry visible through portals (doors and windows) in a three-dimensional model. We present several solutions: images texture-mapped onto the portal, 3D warping of single reference images, and 3D warping of layered depth im...

The paper focuses on solving the problem of hair removal in dermatology applications. The proposed hair removal algorithm is based on Gabor filtering and PDE-based image reconstruction. It also includes the edge sharpening stage using a new warping algorithm. The idea of warping is to move pixels from the neighborhood of the blurred edge closer to the edge. The proposed technique preserves the ...

This works extends the Random Embedding Bayesian Optimization approach by integrating a warping of the high dimensional subspace within the covariance kernel. The proposed warping, that relies on elementary geometric considerations, allows mitigating the drawbacks of the high extrinsic dimensionality while avoiding the algorithm to evaluate points giving redundant information. It also alleviate...

Animated image sequences often exhibit a large amount of inter-frame coherence which standard rendering algorithms and pipelines are ill-equipped to exploit, limiting their efficiency. To address this inefficiency we transfer rendering results across frames using a novel image warping algorithm based on fixed point iteration. We analyze the behavior of the iteration and describe two alternative...