Abstract: A learning device allocates which feature value of a sub-object is extracted by a module from a group of sub-objects constituting an object of an image to each of modules that extract feature values of the object in the image in a deep neural network that is a learning target. After that, the learning device performs first learning to perform learning of the respective modules so that the respective modules are capable of precisely picking up regions of sub-objects allocated to the modules using information indicating the regions of the sub-objects in an image for each of images and second learning to perform learning of the respective modules so that analysis precision of an image analysis is further improved using a result of the image analysis based on the feature values of the sub-objects picked up by the respective modules.

Abstract: An index generation unit (15a) generates an index in which a plurality of combinations of query data that is a collation source and target data that is a collation destination are listed in a predetermined order. A batch generation unit (15b) uses a plurality of combinations of query data and target data in an order according to the index to generate a batch with a predetermined volume. A collation unit (16) calculates a degree of similarity between the query data and the target data for each combination included in the batch, in which processes are parallelized and performed.

Abstract: A learning device allocates which feature value of a sub-object is extracted by a module from a group of sub-objects constituting an object of an image to each of modules that extract feature values of the object in the image in a deep neural network that is a learning target. After that, the learning device performs first learning to perform learning of the respective modules so that the respective modules are capable of precisely picking up regions of sub-objects allocated to the modules using information indicating the regions of the sub-objects in an image for each of images and second learning to perform learning of the respective modules so that analysis precision of an image analysis is further improved using a result of the image analysis based on the feature values of the sub-objects picked up by the respective modules.

Abstract: An index generation unit (15a) generates an index in which a plurality of combinations of query data that is a collation source and target data that is a collation destination are listed in a predetermined order. A batch generation unit (15b) uses a plurality of combinations of query data and target data in an order according to the index to generate a batch with a predetermined volume. A collation unit (16) calculates a degree of similarity between the query data and the target data for each combination included in the batch, in which processes are parallelized and performed.

Abstract: An information processing device (10) predicts a ratio (scale) of features to input data, and, when the predicted scale is equal to or smaller than a predetermined value, divides the input data and outputs the divided data to an analysis device (20). When the predicted scale is significantly smaller than the predetermined value, the information processing device (10) also divides the input data into smaller pieces as the scale is smaller, and outputs the pieces to the analysis device (20). The information processing device (10) also predicts a scale of the features with respect to the input data by machine learning using training data.