Abstract: An image processing device according to the present disclosure generates a second image by executing quantization processing on a first image, generates a feature map based on the second image, generates an existence probability map indicating distribution of existence probability of an object in the first image by using the feature map, performs matching determination processing that is processing of determining whether a value of each cell matches a fixed value for the feature map or the existence probability map, and performs dequantization processing on the feature map or the existence probability map based on a result of the matching determination processing.

Abstract: An image processing device according to the present disclosure generates a feature map based on an image, determines whether a value of each cell of the feature map is equal to or larger than a lower limit threshold value, calculates an activation value by inputting the value of each cell of the feature map equal to or larger than the lower limit threshold value to an activation function, and generates an existence probability map indicating distribution of existence probability of an object in the image by using the activation value.

Abstract: An image processing device according to an embodiment includes: an image division unit dividing input image data into first division image data and second division image data having a predetermined overlap region with the first division image data according to a size of a kernel used for image processing; a reuse data determination unit determining first reuse data reused in performing the image processing to the second division image data among the first division image data; and a memory management unit, in a memory, storing first processed data of a first division image obtained by performing the image processing to the first division image data, in a region other than a region storing the first reuse data, and allocates a region storing the second division image data so as to be adjacent to the region storing the first reuse data.

Abstract: Provided are a device, a method, and a program which allow learning models to be appropriately evaluated or trained. The evaluation device according to an aspect performs the steps of: (A) obtaining, using checking data, a first execution result based on a first learning model as an exemplar model; (B) obtaining, using the checking data, a second execution result based on a second learning model; (C) determining whether or not the first and second execution results satisfy a logical formula; and (D) comparing, using a Bayesian statistical model checking method, respective behaviors of the first and second learning models with each other on the basis of a result of the determination in the step (C).

Abstract: A video decoding processing apparatus which can reduce overhead for the start of parallel decoding processing. The video decoding processing apparatus includes a parsing unit, and first and second video processing units. A coding bit stream including information of largest coding units each having at least a prescribed pixel size is supplied to an input terminal of the parsing unit. The parsing unit performs parsing of the syntax of the coding bit stream to thereby generate parallel-processable first and second intermediate streams from the largest coding unit. The first and second video processing units parallel-process the first and second intermediate streams generated from the parsing unit.

Abstract: The data processor includes CPU operable to execute an instruction included in an instruction set. The instruction set includes a load instruction for reading data on a memory space. The data read according to the load instruction includes data of a format type having a data-read-branching-occurrence bit region. The CPU includes a data-read-branching control register; a data-read-branching address register; and a read-data-analyzing unit. On condition that a bit value showing the occurrence of data read branching has been set on the data-read-branching-occurrence bit region, and a value showing the data-read-branching-occurrence bit remaining valid has been set on the data-read-branching control register, the switching between processes is performed by branching to an address stored in the data-read-branching address register.

Abstract: Disclosed is a video decoding processing apparatus which can reduce overhead for the start of parallel decoding processing. The video decoding processing apparatus includes a parsing unit, and first and second video processing units. A coding bit stream including information of largest coding units each having at least a prescribed pixel size is supplied to an input terminal of the parsing unit. The parsing unit performs parsing of the syntax of the coding bit stream to thereby generate parallel-processable first and second intermediate streams from the largest coding unit. The first and second video processing units parallel-process the first and second intermediate streams generated from the parsing unit.

Abstract: The present invention provides an image encoding device which does not necessitate reference of a quantization parameter between consecutive macroblocks across a parallel processing area boundary without forming slices. The image encoding device encodes a macroblock of an encoding target image by parallel processing sequentially from the top of a parallel processing area, and possesses an encoding element for every parallel processing area. When all the quantized orthogonally-transformed coefficients of a top macroblock of the parallel processing area are zero, the encoding element adds a non-zero coefficient to a part of the coefficients, making the coefficients non-zero. Accordingly, generation of a skip macroblock in the top macroblock of each parallel processing area is suppressed. Since slice formation is not necessary, the prediction over a parallel processing area boundary is applied, and encoding efficiency improves.

Abstract: The data processor includes CPU operable to execute an instruction included in an instruction set. The instruction set includes a load instruction for reading data on a memory space. The data read according to the load instruction includes data of a format type having a data-read-branching-occurrence bit region. The CPU includes a data-read-branching control register; a data-read-branching address register; and a read-data-analyzing unit. On condition that a bit value showing the occurrence of data read branching has been set on the data-read-branching-occurrence bit region, and a value showing the data-read-branching-occurrence bit remaining valid has been set on the data-read-branching control register, the switching between processes is performed by branching to an address stored in the data-read-branching address register.

Abstract: The variable length decoder has a memory device including a plurality of lookup tables, and sequentially decodes codewords of variable-length codes using the memory device. The decoded values corresponding to the codewords and control information pieces are stored in the lookup tables. In decoding one codeword, one lookup table is selected from among the plurality of lookup tables. In the decode, one decoded value corresponding to the one codeword, and a control information piece for selecting a next lookup table depending on the decoded value and used for a next decode are produced from the selected lookup table in response to the one codeword in parallel.

Abstract: The present invention provides an image encoding device which does not necessitate reference of a quantization parameter between consecutive macroblocks across a parallel processing area boundary without forming slices. The image encoding device encodes a macroblock of an encoding target image by parallel processing sequentially from the top of a parallel processing area, and possesses an encoding element for every parallel processing area. When all the quantized orthogonally-transformed coefficients of a top macroblock of the parallel processing area are zero, the encoding element adds a non-zero coefficient to a part of the coefficients, making the coefficients non-zero. Accordingly, generation of a skip macroblock in the top macroblock of each parallel processing area is suppressed. Since slice formation is not necessary, the prediction over a parallel processing area boundary is applied, and encoding efficiency improves.

Abstract: The present invention provides a stream processing apparatus capable of improving the processing performance in the case of continuously processing a plurality of data streams. A control stream, different from a data stream, is prepared, and a program and a parameter are updated in advance in accordance with the control stream. Double buffer areas are prepared in a memory of the stream processing apparatus into which the program and the parameter are stored. The location of the data stream to be input is written in the control stream, and buffers for reading the data stream are multiplexed so as to read in advance the top portion of the data stream to be processed next.