Media Vision

Klics, Inc.

Adrian S Lewis

Gregory P Knowles

A decompression method uses four coefficient inverse perfect reconstruction digital filters. The coefficients of these inverse perfect reconstruction digital filters require a small number of additions to implement thereby enabling rapid decompression in software. The method partially inverse transforms a sub-band decomposition to generate a small low pass component image. This small image is expanded in one dimension by performing interpolation on the rows of the small image and is expanded in a second dimension by replicating rows of the interpolated small image. Transformed chrominance data values may be inverse transformed using inverse perfect reconstruction digital filters having a fewer number of coefficients than the inverse perfect reconstruction digital filters used to inverse transform the corresponding transformed luminance data values. In one embodiment, two coefficient Haar digital filters are used as the inverse perfect reconstruction digital filters which inverse transform transformed chrominance data values. Variable-length tokens are used in the compressed data stream to indicate changes in encoding methods used to encode data values in the compressed data stream.

Data compression and decompression

A compression and decompression method using a wavelet decomposition, frequency based tree encoding, tree based motion encoding, frequency weighted quantization, Huffman encoding, and tree based activity estimation for bit rate control. Forward and inverse quasi-perfect reconstruction transforms are used to generate the wavelet decomposition and to reconstruct data values close to the original data values. The forward and inverse quasi-perfect reconstruction transforms utilize special filters at the boundaries of the data being transformed and/or inverse transformed.

Growing community of inventors

Palma, Spain

Adrian S Lewis