Abstract: Approaches to three-dimensional (3D) data reconstruction are presented. The 3D data comprises 2D images. In some embodiments, the 2D images are directionally interpolated to generate directionally-interpolated 3D data. The directionally-interpolated 3D data are then segmented to generate segmented directionally-interpolated 3D data. The segmented directionally-interpolated 3D data is then meshed. In other embodiments, a 3D data set, which includes 2D flow images, is accessed. The accessed 2D flow images are then directionally interpolated to generate 2D intermediate flow images.

Abstract: A video coding system and method enable communication of a noisy video signal over communications channels, such as low capacity channels (e.g., telephone lines), by efficiently decomposing the video signal into an underlying structure component and a texture component using subband decomposition. Because much of a texture component is random noise, the texture component lends itself to temporal decimation. This predicament results in a texture frame being encoded and transmitted once every D.sub.t frames and allows texture frames to be sent at a lower rate than underlying structure frames. Because the underlying structure component contains most of the information related to structure, it is decimated in the spatial domain, which allows for maximum data transfer.

Abstract: A subband coder with jointly optimized multistage residual quantizers and entropy coders for image coding which allows the subband coder to exploit both linear and non-linear dependencies that may exist within and across the subbands. Both inter- and intra-band dependencies are simultaneously exploited via jointly optimizing the subband coders and entropy coders in an entropy constrained optimization framework involving rate, distortion and complexity. The multistage quantization structure and the effectiveness of the statistical modeling algorithm results in an attractive balance among reproduction quality, rate and complexity.

Abstract: A method and apparatus for the automatic analysis, synthesis and modification of audio signals, based on an overlap-add sinusoidal model, is disclosed. Automatic analysis of amplitude, frequency and phase parameters of the model is achieved using an analysis-by-synthesis procedure which incorporates successive approximation, yielding synthetic waveforms which are very good approximations to the original waveforms and are perceptually identical to the original sounds. A generalized overlap-add sinusoidal model is introduced which can modify audio signals without objectionable artifacts. In addition, a new approach to pitch-scale modification allows for the use of arbitrary spectral envelope estimates and addresses the problems of high-frequency loss and noise amplification encountered with prior art methods.

Abstract: A method and apparatus for the automatic analysis, synthesis and modification of audio signals, based on an overlap-add sinusoidal model, is disclosed. Automatic analysis of amplitude, frequency and phase parameters of the model is achieved using an analysis-by-synthesis procedure which incorporates successive approximation, yielding synthetic waveforms which are very good approximations to the original waveforms and are perceptually identical to the original sounds. A generalized overlap-add sinusoidal model is introduced which can modify audio signals without objectionable artifacts. In addition, a new approach to pitch-scale modification allows for the use of arbitrary spectral envelope estimates and addresses the problems of high-frequency loss and noise amplification encountered with prior art methods.