Abstract: According to the present disclosure, a weight parameter of a neural network is divided into a plurality of portions having a certain size and approximation is individually performed on the portions using a weighted sum of the codebook vectors.

Abstract: There is provided with an information processing apparatus. An acquisition unit acquires data. A detection unit detects a determination target from the data. A training unit performs training of a determiner for performing category determination concerning the determination target using information of the determination target detected by the detection unit and intermediate data being generated during detection processing by the detection unit and being different from the information of the determination target.

Abstract: An apparatus includes a multitask setting unit configured to set a plurality of recognition tasks for which a multilayer neural network is trained, a neural network (NN) training unit configured to train the multilayer NN for the set plurality of recognition tasks, an NN evaluation unit configured to evaluate a training result of the multilayer NN, and a training parameter setting unit configured to set a training parameter in training the multilayer NN for the plurality of recognition tasks, based on a result of evaluation by the NN evaluation unit.

Abstract: There is provided with an information processing apparatus. An acquisition unit acquires data. A detection unit detects a determination target from the data. A training unit performs training of a determiner for performing category determination concerning the determination target using information of the determination target detected by the detection unit and intermediate data being generated during detection processing by the detection unit and being different from the information of the determination target.

Abstract: There is provided with a training apparatus for training a neural network. The neural network, when an input image is inputted, outputs a detection result of a first type and a detection result of a second type for each position of the input image. A training data obtaining unit obtains a training image to be input to the neural network for training. An error map obtaining unit obtains an error map indicating a detection error for a detection result of the first type, for each position of the training image. A training unit trains the neural network using the detection result of the first type and the detection result of the second type that are obtained by inputting the training image to the neural network, and the error map.

Abstract: An apparatus includes a multitask setting unit configured to set a plurality of recognition tasks for which a multilayer neural network is trained, a neural network (NN) training unit configured to train the multilayer NN for the set plurality of recognition tasks, an NN evaluation unit configured to evaluate a training result of the multilayer NN, and a training parameter setting unit configured to set a training parameter in training the multilayer NN for the plurality of recognition tasks, based on a result of evaluation by the NN evaluation unit.

Abstract: An image recognition method for an image recognition apparatus includes detecting, setting, acquiring, selecting, and specifying. At least one part of an identification target is selected from an identification target image. An inquiry region is set based on the detected part. A feature amount of the set inquiry region is acquired. At least one instance image corresponding to the identification target image is selected based on the acquired feature amount. A specific region of the identification target from the identification target image is specified based on the selected instance image.

Abstract: To enable efficient learning a neural network in an adaptive domain, a learning apparatus for learning a multilayer neural network (multilayer NN), comprises: a first learning unit configured to learn a first multilayer NN by using a first data group; a first generation unit configured to generate a second multilayer NN by inserting a conversion unit for performing predetermined processing between a first layer and a second layer following the first layer in the first multilayer NN; and a second learning unit configured to learn the second multilayer NN by using a second data group different in characteristic from the first data group.

Abstract: An image recognition method for an image recognition apparatus includes detecting, setting, acquiring, selecting, and specifying. At least one part of an identification target is selected from an identification target image. An inquiry region is set based on the detected part. A feature amount of the set inquiry region is acquired. At least one instance image corresponding to the identification target image is selected based on the acquired feature amount. A specific region of the identification target from the identification target image is specified based on the selected instance image.

Abstract: An image processing apparatus includes an acquisition unit, a first detection unit, a selection unit, and a correction unit. The acquisition unit acquires an image including a target object having a plurality of parts. The first detection unit detects a candidate region of each of the plurality of parts of the target object included in the acquired image using a previously learned model. The selection unit selects, based on the candidate region detected by the first detection unit, a first part having relatively high reliability and a second part having relatively low reliability from among the plurality of parts. The correction unit corrects the model by changing a position of the second part based on the first part selected by the selection unit.

Abstract: According to the present disclosure, a weight parameter of a neural network is divided into a plurality of portions having a certain size and approximation is individually performed on the portions using a weighted sum of the codebook vectors.

Abstract: It aims to accurately identify a class concerning classification of an object for each area of an image. A plurality of subsets are created from a plurality of learning images, and an area discriminator for performing area discrimination of the learning images is learned for each subset created. Then, area discrimination of a plurality of learning evaluation images is performed by using the plurality of learned area discriminators. Further, an integrated identifier is learned based on an area discrimination result performed to the plurality of learning evaluation images and scene information previously associated with each of the plurality of learning evaluation images.

Abstract: A dictionary for detection of an object is created from an image obtained by performing an image process, which depends on a first image process parameter, on the training image of the detection target object. The dictionary created based on the image process depending on the first image process parameter is determined, based on a result of detecting the object from an image obtained by performing an image process, which depends on the first image process parameter, on a photographed image based on the dictionary. A second image process parameter is determined, based on a result of detecting the object from an image obtained by performing an image process, which depends on the second image process parameter, on the photographed image using the determined dictionary.

Abstract: An image recognition method for an image recognition apparatus includes detecting, setting, acquiring, selecting, and specifying. At least one part of an identification target is selected from an identification target image. An inquiry region is set based on the detected part. A feature amount of the set inquiry region is acquired. At least one instance image corresponding to the identification target image is selected based on the acquired feature amount. A specific region of the identification target from the identification target image is specified based on the selected instance image.

Abstract: An image processing apparatus includes an acquisition unit, a first detection unit, a selection unit, and a correction unit. The acquisition unit acquires an image including a target object having a plurality of parts. The first detection unit detects a candidate region of each of the plurality of parts of the target object included in the acquired image using a previously learned model. The selection unit selects, based on the candidate region detected by the first detection unit, a first part having relatively high reliability and a second part having relatively low reliability from among the plurality of parts. The correction unit corrects the model by changing a position of the second part based on the first part selected by the selection unit.

Abstract: An information processing apparatus, for training a classifier that classifies local regions of an object, includes a feature amount setting unit, a selection unit, and a training unit. The feature amount setting unit sets a feature amount to be used by the classifier. The selection unit selects a local region of the object based on a predetermined selection condition based on positions for obtaining the feature amount set by the feature amount setting unit. The training unit trains the classifier using the feature amount set by the feature amount setting unit and the local region selected by the selection unit.

Abstract: In order to precisely identify a class relating to classification of an object at each of regions of an image, an image processing apparatus includes an acquisition unit configured to acquire image capturing information when the object has been captured, an estimation unit configured to estimate distribution relating to a degree of existence of each class which indicates classification of the object at each of predetermined regions of a captured image of the object based on the image capturing information acquired by the acquisition unit, and an identification unit configured to identify the class at each of the regions based on distribution information indicating the distribution estimated by the estimation unit and image information relating to the captured image.

Abstract: It aims to accurately identify a class concerning classification of an object for each area of an image. A plurality of subsets are created from a plurality of learning images, and an area discriminator for performing area discrimination of the learning images is learned for each subset created. Then, area discrimination of a plurality of learning evaluation images is performed by using the plurality of learned area discriminators. Further, an integrated identifier is learned based on an area discrimination result performed to the plurality of learning evaluation images and scene information previously associated with each of the plurality of learning evaluation images.

Abstract: A plurality of images obtained by capturing a recognition target object from different viewpoint positions is acquired, and a portion set on the recognition target object in the image is received as a set portion for each of the images. A plurality of feature points is set in each of the images so as to set a larger number of feature points at the set portion than at an unset portion other than the set portion. The recognition target object is learned using image feature amounts at the feature points.

Abstract: A probability at which a target object takes a target object state is acquired for each of target object states that the target object is allowed to take, and a distribution of the probabilities is acquired. A success rate is acquired, for each relative target object state being determined in advance for a position and orientation of an image capturing device, at which the target object is successfully identified from a captured image obtained by capturing the target object having the relative target object state, and a distribution of the success rates is acquired. A position and orientation that the device is to take is determined based on the distribution of the success rates acquired for each of a plurality of positions and orientations that the image capturing apparatus is allowed to take, and the distribution of the probabilities.