摘要:
An information processing apparatus and method enables a pattern discriminator to learn. The apparatus establishes a branch structure appropriate for learning a discriminator having the branch structure without increasing processing time. The apparatus includes a preliminary learning unit to learn a preliminary discriminator for a respective one of a plurality of combinations of variations in variation categories in a discrimination target pattern. A branch structure determination unit is provided to perform discrimination processing using the preliminary discriminator and to determine a branch structure of a main discriminator based on a result of the discrimination processing. A main learning unit is included to learn the main discriminator based on the branch structure.
摘要:
Even when a local area is varied, degradation in recognition accuracy and detection accuracy is suppressed. To that end, a pattern processing apparatus includes a reference local area setting portion 1802 for setting a reference local area based on the detection result of a feature point by a face organ feature point detecting portion 101, a varied local area generating portion 1803 for generating a plurality of varied local area patterns by referring to an image area near the reference local area, a similarity calculating portion 106 for calculating similarities in the reference local areas and in the varied local area patterns between the input pattern and the registered pattern, a representative similarity calculating portion 107 for calculating representative similarity from among the similarities, and a classifying portion 109 for determining a class to which the input pattern belongs.
摘要:
An image processing apparatus classifies a variation of a target object included in an image from a specific state as one of a plurality of types of attributes, and holds, for each variation attribute, a correction parameter for spatial transformation that corrects the target object to the specific state. The image processing apparatus generates an input image vector by vectorizing at least a partial region of the input image, and determines a variation attribute by detecting a variation of the target object from the specific state in the input image. Then, the image processing apparatus generates a transformed image vector by performing the spatial transformation on the input image vector, using a correction parameter selected based on the determined variation attribute from among the correction parameters held for respective variation attributes.
摘要:
In a convolution operation circuit, a first and a second shift registers provide data to a first and a second inputs of a plurality of multipliers, a first and a second storage units store data to be supplied to the first and the second shift registers, a plurality of cumulative adders accumulate output from the plurality of multipliers, a third storage unit latches output from the plurality of cumulative adders at predetermined timing, a fourth storage unit stores data to be stored in the first and the second storage units and data output from the third storage unit, and a control unit sets data stored in the first and the second storage units to the first and the second shift registers at predetermined timing, causes the first and the second shift registers to perform shift operations in synchronization with an operation of the cumulative adder.
摘要:
A pattern identification unit generation method of generating a pattern identification unit in which a weak discriminator array obtained by cascade-connecting a plurality of weak discriminators branches, and weak discriminator arrays are connected to respective arms after branching, evaluates based on a processing result obtained by inputting a set of evaluation data to the weak discriminator array whether or not a weak discriminator array after branching reaches the number of stages to be connected. The number of stages of weak discriminators to be connected without branching as the weak discriminator array is determined based on this evaluation result.
摘要:
A calculation processing apparatus, which executes network calculations by a hierarchical network configured by connecting a plurality of logical processing nodes that execute calculations for input data to generate calculation result data, calculates memory sizes of buffers required for the network calculations, based on the configuration of the network calculations, in association with a plurality of types of buffer assignment methods for assigning storage areas for intermediate buffers used to hold calculation result data in correspondence with the plurality of processing nodes which configure the network, and executes the network calculations using buffers assigned by the buffer assignment method, which is selected based on the calculated memory size, in the execution sequence according to that buffer assignment method.
摘要:
A target object gripping apparatus comprises: an estimation unit configured to estimate an orientation of a target object based on orientation estimation parameters; a gripping unit configured to grip the target object based on the orientation of the target object estimated by the estimation unit; a detection unit configured to detect a failure of gripping by the gripping unit; and a modifying unit configured to modify the orientation estimation parameters based on the orientation of the target object when the detection unit detects a gripping failure.
摘要:
In order to reduce wasteful operation processes by operation units, a data supply unit reads out a plurality of pieces of pixel data from an image memory and temporarily stores the pixel data in registers, and outputs the pixel data while shifting the pixel data in one direction; product-sum operators concurrently perform operation processes based on multiple pieces of pixel data outputted from the data supply unit and filter coefficients in a filter kernel. At this time, a selector within the data supply unit selects the pixel data stored in each register in accordance with the type of the filter operation process, and outputs the selected pixel data to the product-sum operators.
摘要:
An information processing apparatus includes a preliminary learning unit configured to learn a preliminary discriminator for a respective one of a plurality of combinations of variations in variation categories in a discrimination target pattern, a branch structure determination unit configured to perform discrimination processing using the preliminary discriminator and to determine a branch structure of a main discriminator based on a result of the discrimination processing, and a main learning unit configured to learn the main discriminator based on the branch structure.