Embodiments described herein provided for an instruction and associated logic to enable a processing resource including a tensor accelerator to perform optimized computation of sparse submatrix operations. One embodiment provides hardware logic to apply a numerical transform to matrix data to increase the sparsity of the data. Increasing the sparsity may result in a higher compression ratio when the matrix data is compressed.

The present technology relates to an image processing apparatus and method that are capable of enhancing encoding efficiency while suppressing a decrease in the efficiency of encoding processing. The image processing apparatus includes an encoding mode setter that sets, in units of coding units having a hierarchical structure, whether a non-compression mode is to be selected as an encoding mode for encoding image data, the non-compression mode being an encoding mode in which the image data is output as encoded data, and an encoder that encodes the image data in units of the coding units in accordance with a mode set by the encoding mode setter. The present disclosure can be applied to, for example, an image processing apparatus.

The present invention relates to an improved method and apparatus for image compression and particularly to an improved block coding apparatus and method for compression for use with the JPEG2000 standard, although not limited to this. Methods for coding and decoding blocks and subbands samples derived from still images video frames or related media, involving three bit-streams and the partitioning of samples from the blocking to define groups, is provided. A first bit-stream encodes the significance of whole groups. A second bit-stream encodes the significance of individual samples within each group. The second bit-stream also encodes an unsigned residual value for each significant group. A third bit stream provides a sign bit and any additional magnitude bits required to represent the significant sample values. Exponent predictors are computal using both exponent bounds and the additional magnitude bits associated with previous samples in the block.

An electronic apparatus and method for adaptive sub-band based coding of hierarchical transform coefficients of a 3D point cloud, is provided. The electronic apparatus stores the 3D point cloud and generates a plurality of voxels from the 3D point cloud. The electronic apparatus generates a plurality of hierarchical transform coefficients by application of a hierarchical transform scheme on the generated plurality of voxels and classifies the plurality of hierarchical transform coefficients into a plurality of sub-bands of hierarchical transform coefficients. The plurality of hierarchical transform coefficients are classified based on a weight of each of the plurality of hierarchical transform coefficients. The electronic apparatus further generates a plurality of quantized levels by application of a different quantization scheme of a set of different quantization schemes on each sub-band of hierarchical transform coefficients of the plurality of sub-bands of hierarchical transform coefficients and encodes the generated plurality of quantized levels.

A method, computer system, and computer-readable medium are provided for point cloud attribute coding by at least one processor. Data associated with a point cloud is received. The received data is transformed through a lifting decomposition based on enabling a scalable coding of attributes associated with the lifting decomposition. The point cloud is reconstructed based on the transformed data.

A method, apparatus, and computer-readable medium for point cloud coefficient coding are provided. Transform coefficients associated with point cloud data are decomposed into set-index values and symbol-index values, the symbol index-value specifying location of the transform coefficient within a set. The decomposed transform coefficients are partitioned into one or more sets based on the set-index values and the symbol-index values. The set-index values of the partitioned transform coefficients are entropy-coded, and the symbol-index values of the partitioned transform coefficients are bypass-coded. The point cloud data is compressed based on the entropy-coded symbol-index values and the bypass-coded set-index values.

In one example, the present disclosure describes a device, computer-readable medium, and method for enabling secure video sharing by exploiting data sparsity. In one example, the method includes applying a transformation to a video dataset containing a plurality of video samples, to produce a plurality of sparse vectors in a first dimensional space, wherein each sparse vector of the plurality of sparse vectors corresponds to one video sample of the plurality of video samples, and multiplying each sparse vector of the plurality of sparse vectors by a transformation matrix to produce a plurality of reduced vectors in a second dimensional space, wherein the dimension of the second dimensional space is smaller than a dimension of the first dimensional space, and wherein the plurality of reduced vectors in the second dimensional space hides information about the video dataset while preserving relational properties between the plurality of video samples.

A method is disclosed including: receiving raw image data corresponding to a series of raw images; processing the raw image data with an encoder to generate encoded data, where the encoder is characterized by an input/output transformation that substantially mimics the input/output transformation of one or more retinal cells of a vertebrate retina; and applying a first machine vision algorithm to data generated based at least in part on the encoded data.

Various embodiments are generally directed to an apparatus, method and other techniques for storing, in memory, at least one master forward transform matrix comprising signed constants having a defined number of precision bits and a sign bit and determining, by processing circuitry, which forward transform matrix to use to perform a transformation based on at least a transform unit size. Further, various techniques may include performing, by the processing circuitry, the transformation on residuals of pixel values of a frame using one of the at least one master forward transform matrix or a forward transform matrix derived from one of the master forward transform matrix at least partially based on the determination.

A method of processing digital images, the method comprising: creating a downsampled version of a digital image by downsampling the digital image; upsampling the downsampled version to create an upsampled version; comparing the digital image to the upsampled version to create residual values describing differences between the digital image and the upsampled version. Applying a wavelet transform to the residual values to create transformed residual values.