Avago Technologies General IP (singapore) Pte. Ltd.

Lsi Corporation

Huawei Technologies Co., Limited

Dmitry Nicolaevich Babin

Ivan Leonidovich Mazurenko

Denis Vladimirovich Parkhomenko

Alexander Alexandrovich Petyushko

Denis Vladimirovich Zaytsev

Denis Vasilyevich Parfenov

Aleksey Alexandrovich Letunovskiy

Pavel Aleksandrovich Aliseychik

Alexander Borisovich Kholodenko

Denis Vasilievich Parfenov

Pavel Aleksandrovich Aliseitchik

Yang Han

Barrett J Brickner

Kun Guan

David G Shaw

Alexander Nikolaevich Filippov

Stanislav Vladimirovich Moiseev

A method of similarity-based deduplication comprising the steps of: receiving an input data block; computing discrete wavelet transform (DWT) coefficients; extracting feature-related DWT data from the computed DWT coefficients; applying quantization to the extracted feature-related DWT data to obtain keys as results of the quantization; constructing a locality-sensitive fingerprint of the input data block; computing a similarity degree between the locality-sensitive fingerprint of the input data block and a locality-sensitive fingerprint of each data block in the plurality of the data blocks in a cache memory; selecting an optimal reference data block as the data block; determining a differential compression is required to be applied based on the similarity degree between the input data block and the optimal reference data block; applying the differential compression to the input data block and the optimal reference data block.

Method and system of similarity-based deduplication

An image processing system comprises an image processor configured to construct a designated functional based on a plurality of functions each associated with a corresponding portion of image information relating to at least first and second images, and to generate a target image utilizing the constructed functional. For example, the functions may comprise a set of functions &#x192;(A), &#x192;(A), . . . , &#x192;(A) of pixels from respective input images A, A, Aof the image information, and the functional may be a function F(X) of the set of functions &#x192;,(A) &#x192;(A), &#x192;(A) where X denotes the target image and is generated by minimizing the functional F(X). The input images may be received from one or more image sources and the target image may be provided to one or more image destinations.

Target image generation utilizing a functional based on functions of information from other images

An image processing system comprises an image processor having image processing circuitry and an associated memory. The image processor is configured to implement a foreground processing module utilizing the image processing circuitry and the memory. The foreground processing module is configured to obtain one or more images, to estimate a foreground region of interest from the one or more images, to determine a plurality of segments of the foreground region of interest, to calculate amplitude statistics for respective ones of the plurality of segments, to classify respective segments as being respective portions of static foreground objects or as being respective portions of dynamic foreground objects based at least in part on the calculated amplitude statistics and one or more defined patterns for known static and dynamic objects, and to remove one or more segments classified as static foreground objects from the foreground region of interest.

Image processor configured for efficient estimation and elimination of foreground information in images

An image processing system comprises an image processor configured to identify edges in an image, to apply a first type of filtering operation to portions of the image associated with the edges, and to apply a second type of filtering operation to one or more other portions of the image. By way of example only, in a given embodiment a clustering operation is applied to the image to identify a plurality of clusters, a first set of edges comprising edges of the clusters is identified, an edge detection operation is applied to the image to identify a second set of edges, a third set of edges is identified based on the first and second sets of edges, and the first type of filtering operation is applied to portions of the image associated with one or more edges of the third set of edges.

Image processor with edge-preserving noise suppression functionality

An image processing system comprises an image processor configured to obtain a first image stream having a first frame rate and a second image stream having a second frame rate lower than the first frame rate, to recover additional frames for the second image stream based on existing frames of the first and second image streams, and to utilize the additional frames to provide an increased frame rate for the second image stream. Recovering additional frames for the second image stream based on existing frames of the first and second image streams illustratively comprises determining sets of one or more additional frames for insertion between respective pairs of consecutive existing frames in the second image stream in respective iterations.

Method and apparatus for increasing frame rate of an image stream using at least one higher frame rate image stream

A method and system for key frame based region of interest (ROI) tracking is disclosed. The method includes storing a key ROI set in a key ROI buffer, the key ROI set including at least one key ROI; designating one of the key ROI in the key ROI set as an active key ROI; receiving a point cloud representing a particular ROI to be processed for tracking; establishing a correspondence between that particular ROI and the active key ROI; determining whether to switch the active key designation to another key ROI in the key ROI set and switching the active key designation accordingly; and determining whether to modify the key ROI set and modifying the key ROI set accordingly.

Three-dimensional region of interest tracking based on key frame matching

An image processing system comprises an image processor configured to perform first and second edge detection operations on respective first and second images to obtain respective first and second edge images, to apply a joint edge weighting operation using edges from the first and second edge images, to generate an edge mask based on results of the edge weighting operation, to utilize the edge mask to obtain a third edge image, and to generate a third image based on the third edge image. By way of example only, in a given embodiment the first image may comprise a first depth image generated by a depth imager, the second image may comprise a two-dimensional image of substantially the same scene as the first image, and the third image may comprise an enhanced depth image having enhanced edge quality relative to the first depth image.

Method and apparatus for image enhancement and edge verification using at least one additional image

A method for estimating error rates in low-density parity check codes includes calibrating an encoder according to specific channel parameters and according to dominant error events in the low-density parity-check code. Dominant codewords are classified based on characteristics of each codeword that are likely to produce similar error rates at similar noise levels; codeword classes that produce the highest error rate are then tested. Error boundary distance is estimated using multiple binary searches on segments. Segments are defined based on codeword, trapping set and biasing noise components of the channel. To improve calculation speed the most significant subclasses of codewords, trapping sets and noise signals are used.

Test signal generator for low-density parity-check decoder

A method for ordering trapping sets to find one or more dominant trapping sets includes analyzing a trapping set and a random set of codewords to generate a distance value for each trapping set, and ordering the trapping sets by the distance value. Distance values may be determined for each trapping set by tracking a vote count wherein a correct decode at a certain noise level produces a 'right' vote and an incorrect decode at a certain noise level produces a 'left' vote. A certain threshold number of &#x201c;left&#x201d; votes terminates processing at that noise level.

Efficient way to construct LDPC code by comparing error events using a voting based method

A method for double precision approximation of a single precision operation is disclosed. The method may include steps (A) to (B). Step (A) may store an input value in a processor. The processor generally implements a plurality of first operations in hardware. Each first operation may receive a first variable as an argument. The first variable may be implemented in a fixed point format at a single precision. The input value may be implemented in the fixed point format at a double precision. Step (B) may generate an output value by emulating a selected one of the first operations using the input value as the argument. The emulation may utilize the selected first operation in hardware. The output value may be implemented in the fixed point format at the double precision. The emulation is generally performed by a plurality of instructions executed by the processor.

Double precision approximation of a single precision operation

An apparatus generally having a first circuit and a second circuit is disclosed. The first circuit may be configured to synthesize a first vector by filtering a second vector based on a third vector. The second circuit may be configured to (i) generate a gain corresponding to a fourth vector, (ii) compare the gain to a plurality of thresholds and (iii) update the third vector as a function of the gain where the compare determines that the gain is not between the thresholds. The fourth vector may be received from a network as an echo of the second vector.

Fast echo gain change detection

An apparatus generally having a first circuit and a second circuit is disclosed. The first circuit may be configured to generate a first sample by filtering an input vector based on (a) a filter vector and (b) a stochastic vector. Each of a plurality of components in the stochastic vector generally has a respective random value. The first circuit may also be configured to generate a second sample as a difference between a third sample and the first sample. The third sample may be received from a network as an echo. The second circuit may be configured to update a subset of a plurality of taps of the filtering where a corresponding one of the components of the stochastic vector has a first value of the random values.

Growing community of inventors

Moscow, Russia

Dmitry Nicolaevich Babin