Akira Nakamura

Bio: Akira Nakamura is an academic researcher from Sony Broadcast & Professional Research Laboratories. The author has contributed to research in topics: Feature detection (computer vision) & Feature (computer vision). The author has an hindex of 16, co-authored 68 publications receiving 660 citations.

Semiconductor laser apparatus for optical heads

Abstract: In a semiconductor laser apparatus for an optical head, a beam splitter having a first surface incline at a predetermined angle with respect to a radiant surface of a semiconductor laser chip is fixed on a photodetector to reduce the cost of the optical head.

Image processing system, learning device and method, and program

Abstract: There are provided an image processing system, a learning device and a method, and a program capable of easily extracting a characteristic amount used for recognition processing. A characteristic point is extracted from a learning model image. According to the characteristic point, a characteristic amount is extracted. The characteristic amount is registered in a leaning model dictionary registration unit (23). Similarly, a characteristic point is extracted from a leaning input image containing a model object contained in the learning model image. According to the characteristic point, a characteristic amount is extracted. The characteristic amount is compared to the characteristic amount registered in the learning model registration unit (23). As the comparison result, the characteristic amount which has become a pair most frequently is registered as a characteristic amount used for recognition processing in a model dictionary registration unit (12). The present invention may be applied to a robot.

Object recognition device, object recognition method, program for object recognition method, and recording medium having recorded thereon program for object recognition method

Abstract: An object recognition device includes: a model image processing unit having a feature point set decision unit setting a feature point set in a model image, and detecting the feature quantity of the feature point set, and a segmentation unit segmenting the model image; a processing-target image processing unit having a feature point setting unit setting a feature point in a processing-target image and detecting the feature quantity of the feature point; a matching unit comparing the feature quantities of the feature points set in the model image and in the processing-target image so as to detect the feature point corresponding to the feature point set, and executes a matching; and a determination unit determining the processing result in the matching unit so as to determine presence/absence of a model object in the processing-target image.

Object Recognition Device, Object Recognition Method, Object Recognition Program, Feature Registration Device, Feature Registration Method, and Feature Registration Program

Abstract: Disclosed herein is an object recognition device for recognition of an object that exists in an image, the device including: an edge image creator configured to create an edge image from a target image; a local feature extractor configured to extract a feature of each edge point on an edge image, and determine a base point and a support point from edge points in a local area defined for the edge image; a feature matching unit configured to implement, with reference to a memory, feature matching of a base point and a support point to thereby acquire a matching pair; and a matching pair determiner configured to determine final matching pairs by employing matching pairs having a highest matching cost among matching pairs obtained by the feature matching unit, wherein an object in a target image is identified based on matching pairs determined by the matching pair determiner.

Differentiable Quantization of Deep Neural Networks.

Abstract: We propose differentiable quantization (DQ) for efficient deep neural network (DNN) inference where gradient descent is used to learn the quantizer's step size, dynamic range and bitwidth. Training with differentiable quantizers brings two main benefits: first, DQ does not introduce hyperparameters; second, we can learn for each layer a different step size, dynamic range and bitwidth. Our experiments show that DNNs with heterogeneous and learned bitwidth yield better performance than DNNs with a homogeneous one. Further, we show that there is one natural DQ parametrization especially well suited for training. We confirm our findings with experiments on CIFAR-10 and ImageNet and we obtain quantized DNNs with learned quantization parameters achieving state-of-the-art performance.

Information processing apparatus, information processing method and program

Abstract: A method is provided for modifying an image. The method comprises displaying an image, the image comprising a portion of an object; and determining if an edge of the object is in a location within the portion. The method further comprises detecting movement, in a member direction, of an operating member with respect to the edge. The method still further comprises moving, if the edge is not in the location, the object in an object direction corresponding to the detected movement; and modifying, if the edge is in the location, the image in response to the detected movement, the modified image comprising the edge in the location.

Passive touch system and method of detecting user input

Abstract: A passive touch system includes a passive touch surface and at least two cameras associated with the touch surface. The at least two cameras acquire images of the touch surface from different locations and have overlapping fields of view. A processor receives and processes images acquired by the at least two cameras to detect the existence of a pointer therein and to determine the location of the pointer relative to the touch surface. Actual pointer contact with the touch surface and pointer hover above the touch surface can be determined.

Interactive user interface for capturing a document in an image signal

Abstract: Devices, methods, and software are disclosed for an interactive user interface for capturing a frame of image data having a representation of a feature. In an illustrative embodiment, a device includes an imaging subsystem, one or more memory components, and one or more processors. The imaging subsystem is capable of providing image data representative of light incident on said imaging subsystem. The one or more memory components include at least a first memory component operatively capable of storing an input frame of the image data. The one or more processors may be enabled for performing various steps. One step may include receiving the image data from the first memory component. Another step may include attempting to identify linear features defining a candidate quadrilateral form in the image data. Another step may include providing user-perceptible hints for guiding a user to alter positioning of the device to enhance a capability for identifying the linear features defining a candidate quadrilateral form in the image data.

System for capturing a document in an image signal

Abstract: Devices, methods, and software are disclosed for capturing a frame of image data having a representation of a feature. In an illustrative embodiment, a device includes an imaging subsystem, one or more memory components, and a processor. The imaging subsystem is capable of providing image data representative of light incident on said imaging subsystem. The one or more memory components include at least a first memory component operatively capable of storing an input frame of the image data. The processor is in communicative connection with executable instructions for enabling the processor for various steps. One step includes receiving the input frame from the first memory component. Another step includes generating a reduced resolution frame based on the input frame, the reduced resolution frame comprising fewer pixels than the input frame, in which a pixel in the reduced resolution frame combines information from two or more pixels in the input frame. Another step includes attempting to identify transition pairs comprising pairs of adjacent pixels in the reduced resolution frame having differences between the pixels that exceed a pixel transition threshold. Another step includes attempting to identify one or more linear features between two or more identified transition pairs in the reduced resolution frame. Another step includes providing an indication of one or more identified linear features in the reduced resolution frame.

Zoom lens and imaging apparatus

Abstract: An object of the present invention is to provide a low-cost, miniature, wide-angle high-zoom-ratio zoom lens that has high imaging performance, and an imaging apparatus equipped with the zoom lens. To achieve the object, a zoom lens including a first lens group having positive refracting power, a second lens group having negative refracting power and a third lens group having positive refracting power in order from an object side in which a lens group P having positive refracting power is arranged closer to an image plane side than the third lens group.