Abstract: To enable a non-defective product image or a defective product image to be easily additionally learned in both a setting mode and a run mode. Designation as to whether an inspection target image displayed on a run mode screen is additionally learned as a non-defective product image or a defective product image is received. A distinguishing device generating unit additionally learns, as a non-defective product image or a defective product image, the inspection target image designated as the additional learning target and updates a distinguishing device.

Abstract: A method of generating a patient-specific prosthesis includes receiving anatomic imaging data representative of a portion of a patient's anatomy. A first digital representation of the anatomic imaging data is defined. The first digital representation of the anatomic imaging data is modified. A second digital representation of the portion of the patient's anatomy is defined based on the modifying of the first digital representation of the anatomic imaging data. A patient-specific prosthesis is generated based at least in part on the second digital representation of the anatomic imaging data.

Abstract: An approach is provided that captures a set of sequential images of an area where there is a selected moving object. Both a keypoint-based (KP-based) matching model and a neural network based (NN-based) matching model are used with the KP-based matching model analyzing most or all of the captured images and the NN-based model being more computational intensive and analyzing a subset of the images. When the KP-based matching model fails to identify the selected object in an image, the NN-based model is used to find the object so that the KP-based matching model can re-establish tracking of the object.

Abstract: Methods and systems perform incremental learning object detection in images and/or videos without catastrophic forgetting of previously-learned object classes. A two-stage neural network object detector is trained to locate and identify objects pertaining to an additional object class by iteratively updating the two-stage neural network object detector until an overall detection accuracy criterion is met. The updating is performed so as to balance minimizing a loss of an initial ability to locate and identify objects pertaining to the previously-learned object classes and maximizing an ability to additionally locate and identify the objects pertaining to the additional object class.

Abstract: A non-transitory computer-readable recording medium storing therein an anomalousness determination program that causes a computer to execute a process includes: sensing a region in an object in each of a plurality of ultrasound examination images using an object sensing technique; based on a result of the sensing and a structure of the object, acquiring a result of sensing each of a plurality of regions in the object in each of the ultrasound examination images; and determining anomalousness in the object based on the result of sensing each of the regions in the ultrasound examination images.

Abstract: A method for predicting spatial positions of several key points on a human body in the near future in an egocentric setting is described. The method includes generating a frame-level supervision for human poses. The method also includes suppressing noise and filling missing joints of the human body using a pose completion module. The method further includes splitting the poses into a global stream and a local stream. Furthermore, the method includes combining the global stream and the local stream to forecast future human locomotion.

Abstract: Techniques are described for assisted creation of video rules via scene analysis. In some implementations, a scene is obtained, a shape of an element in a ground plane shown in the images of the scene is identified, user input that defines a shape of a region of interest used in a video rule is obtained, that the shape of the region satisfies a reshape criteria based on the shape of the element in the ground plane shown in the images of the scene is determined, and a reshaped region based on the shape of the element in the ground plane shown in the images of the scene is determined.

Abstract: The present disclosure relates to a computer-implemented method for generating an enhanced image from an original image, the method including segmenting the original image into a segmented image using an artificial neural network; curve fitting the segmented image to determine boundary artifacts; removing the determined boundary artifacts to generate a smoothed boundary image; and generating the enhanced image from the original image and the smoothed boundary image. The image maybe enhanced further by correcting for glare and adding artificial light.

Abstract: A computer-implemented method for classifying a document type of a document in an image and extracting data from the classified document comprising acquiring image data that comprises data relating to at least a part of the document. Textual classification of the document image is then attempted by machine recognition of textual characters to obtain classification data; and using the classification data to classify the document in the image.

Abstract: In one embodiment, a set of bounding box candidates are plotted onto a 2D space based on their respective dimension (e.g., widths and heights). The bounding box candidates are clustered on the 2D space based on the distribution density of the bounding box candidates. For each of the clusters of the bounding box candidates, an anchor box is determined to represent the corresponding cluster. A neural network model is trained based on the anchor boxes representing the clusters. The neural network model is utilized to detect or recognize objects based on images and/or point clouds captured by a sensor (e.g., camera, LIDAR, and/or RADAR) of an autonomous driving vehicle.

Abstract: A method of semiconductor-wafer image alignment is performed at a semiconductor-wafer defect-inspection system. In the method, a semiconductor wafer is loaded into the semiconductor-wafer defect-inspection system. Pre-inspection alignment is performed for the semiconductor wafer. After performing the pre-inspection alignment, a first swath is executed to generate a first image of a first region on the semiconductor wafer. An offset of a target structure in the first image with respect to a known point is determined. Defect identification is performed for the first image, using the offset. After executing the first swath and determining the offset, a second swath is executed to generate a second image of a second region on the semiconductor wafer. While executing the second swath, run-time alignment of the semiconductor wafer is performed using the offset.

Abstract: Alias capture to support searching for an object-of-interest is disclosed. A method includes capturing, using a camera with a defined field of view, video image frames that include a moving object-of-interest. The method also includes tracking the object-of-interest over a period of time starting when the object-of-interest enters the field of view and ending when the object-of-interest exits the field of view. The method also includes detecting, at a point in time in-between the start and end of the period of time of the tracking, a threshold exceeding change in an appearance of the object-of-interest. The method also includes creating, before the end of the period of time of the tracking, a new object profile for the object-of-interest in response to the detecting of the threshold exceeding change.

Abstract: A method of searching for similar images is performed at a computing system having one or more processors and memory. The method includes receiving an input image having content, and generating a feature vector corresponding to the input image according to a trained classifier model. The feature vector has multiple components. The method further includes encoding the feature vector as a similarity hash by quantizing each component. The method also includes, for each reference image in a plurality of reference images: obtaining a reference hash for the reference image; computing similarity between the input image and the reference image by computing a distance between the reference hash and the similarity hash; and determining whether the computed distance is within a predetermined threshold. When the computed distance is within the predetermined threshold, the computing system returns the reference image as an image that is similar to the input image.

Abstract: A center of gravity of a chest front image is obtained by a center of gravity calculation using pixel values in the chest front image, and a vicinity pixel position of the center of gravity is set as a center position P of a circular search region having a radius of a predetermined length. The shortest line segment of four line segments obtained by drawing perpendicular lines down from the center position P of an irradiation field to boundary lines of the upper, lower, left and right irradiation fields. The line segment, which is a radius of the circular search region, is rotationally displaced for each predetermined angle ?, and the profile of the average value for each predetermined angle ? is obtained. From the displacement angle ? of the line segment in which the average value becomes minimum, the orientation of the subject from the head to the abdomen reflected in the chest front image is determined.

Abstract: Technologies for thermal enhanced semantic segmentation include a computing device having a visible light camera and an infrared camera. The computing device receives a visible light image of a scene from the visible light camera and an infrared image of the scene from the infrared camera. The computing device registers the infrared image to the visible light image to generate a registered image. Registering the infrared image may include increasing resolution of the infrared image. The computing device generates a thermal boundary saliency image based on the registered infrared image. The computing device may generate the thermal boundary saliency image by applying a Gabor jet convolution to the registered infrared image. The computing device performs semantic segmentation on the visible light image, the registered infrared image, and the thermal boundary saliency image to generate a pixelwise semantic classification of the scene. Other embodiments are described and claimed.

Abstract: A motion detecting section 30 detects the attitude of a head-mounted display device worn on the head of a user. A visual line direction determining section 32 determines a visual line direction in accordance with the detected attitude of the head-mounted display device. An image generating section 34 generates an image based on the determined visual line direction. An image providing section 36 provides the head-mounted display device with the generated image. An instruction acquiring section 24 acquires from an input device 6 an instruction to switch the visual line direction. When the instruction acquiring section 24 acquires the switching instruction, the visual line direction determining section 32 changes the visual line direction by a predetermined angle.

Abstract: A method and device for generating a composite group image from subgroup images is provided. Subgroup images, each having a common background, are accessed. The boundaries of a subgroup area within each of the subgroup images is determined. At least one horizontal and at least one vertical shift factor is determined using the determined boundaries. An arrangement for the subgroup images based on the at least one horizontal and the at least one vertical shift factor is generated. The composite group image is generated by blending the subgroup images arranged in the arrangement.

Abstract: Disclosed is a method for detecting pedestrians in an image by using Gaussian penalty. Initial pedestrian boundary box is screened using a Gaussian penalty, to improve the pedestrian detection performance, especially sheltered pedestrians in an image. The method includes acquiring a training data set, a test data set and pedestrian labels of a pedestrian detection image; using the training data set for training to obtain a detection model by using a pedestrian detection method, and acquiring initial pedestrian boundary box and confidence degrees and coordinates thereof; performing Gaussian penalty on the confidence degrees of the pedestrian boundary box, to obtain confidence degree of the pedestrian boundary box after the penalty; and obtaining final pedestrian boundary boxes by screening the pedestrian boundary boxes. Thus, repeated boundary boxes of a single pedestrian are removed while reserving boundary boxes of sheltered pedestrians, thereby realizing the detection of the pedestrians in an image.

Abstract: System and method for tracking at least one device in real-time. A position encoding unit is configured to detect a position of the device, which is configured to move in a first direction. A stage is configured to selectively move with a forward axis of travel and a reverse axis of travel substantially parallel to the first direction and having a stage speed controllable by a controller. A ranged imaging unit is fixedly mounted to the stage and may selectively scan the device upon command. The controller is configured to obtain a speed of the device. When the device reaches a pre-programmed position, the stage is selectively moved at the stage speed synchronized to the speed of the device to within a predefined match value. The ranged imaging unit is employed to determine dimensional characteristics of the device.

Abstract: An example apparatus includes a feature extractor to generate a first image descriptor based on a first image of a first retail product tag corresponding to a first category, the first image descriptor representative of one or more visual features of the first retail product tag; a feature descriptor generator to generate a feature descriptor corresponding to the first retail product tag by concatenating the first image descriptor and a first category signature corresponding to the first retailer category; and a classifier to generate a first probability value corresponding to a first type of promotional product tag and a second probability value corresponding to a second type of promotional product tag based on the feature descriptor; and determine whether the first retail product tag corresponds to the first type of promotional product tag or the second type of promotional product tag based on the first and second probability values.