VI. IMAGE PROCESSING AND COMMUNICATION

The objective of the work in this area is to take advantage of the perceptual, multiscale image region structure, in conjunction with more classical image and video processing techniques, to improve the quality and efficiency of visual communication. Both spatial and transform (Fourier, Wavelet) domain representations are used as needed. Within the spatial domain, both (region) structure based and purely signal (pixel) level descriptions are used. Following are the main categories in which work has been done:

VI.1. Video Compression using Wyner-Ziv Codes: Predictive coding is posed as a variant of the Wyner-Ziv coding, and problems in source and channel coding of video are addressed in this framework.

VI.2. Multiscale structure based image representation using a set of regions: (i) of appropriate granularity for best image compression, (ii) of appropriately rescaled size for image magnification or superresolution, and (iii) for smoothing for image quality restoration through structure-preserving denoising.

VI.3. Multiscale structure based video compression, by estimating and coding region motion instead of pixel motion.

VI.4. Block-based motion estimation for missing video frame interpolation, and spatially scalable (multiresolution) video coding.

VI.5. Computationally efficient transform domain approaches to image and video denoising, changing resolution and watermarking, both using uncompressed data and directly on the compressed data.

VI.6. Efficient spatiotemporal filtering for video denoising.

VI.7. Detection of photometric distribution discontinuities in video to locate shot changes.