Abstract: Systems and methods for automating image annotations are provided, such that a large-scale annotated image collection may be efficiently generated for use in machine learning applications. In some aspects, a mobile device may capture image frames, identifying items appearing in the image frames and detect objects in three-dimensional space across those image frames. Cropped images may be created as associated with each item, which may then be correlated to the detected objects. A unique identifier may then be captured that is associated with the detected object, and labels are automatically applied to the cropped images based on data associated with that unique identifier. In some contexts, images of products carried by a retailer may be captured, and item data may be associated with such images based on that retailer's item taxonomy, for later classification of other/future products.

Abstract: An apparatus for inputting information into a computer includes a 3d sensor that senses 3d information and produces a 3d output. The apparatus includes a 2d sensor that senses 2d information and produces a 2d output The apparatus includes a processing unit which receives the 2d and 3d output and produces a combined output that is a function of the 2d and 3d output. A method for inputting information into a computer. The method includes the steps of producing a 3d output with a 3d sensor that senses 3d information. There is the step of producing a 2d output with a 2d sensor that senses 2d information. There is the step of receiving the 2d and 3d output at a processing unit. There is the step of producing a combined output with the processing unit that is a function of the 2d and 3d output.

Abstract: A system for processing breast tissue images includes an image processing computer and a user interface operatively coupled to the image processing computer, wherein the image processing computer is configured to obtain image data of breast tissue, processing the image data to generate a set of reconstructed image slices, the reconstructed image slices collectively depicting the breast tissue, process respective subsets of the reconstructed image slices to generate a set of image slabs, each image slab comprising a synthesized 2D image of a portion of the breast tissue obtained from a respective subset of the set of reconstructed image slices.

Abstract: Provided are a video data acquisition unit that acquires video data; a parking detection unit that detects whether the vehicle is parked; a moving body detection unit that detects a moving body from video data while the vehicle is parked; an acceleration detection unit that detects acceleration applied to the vehicle; a parking position determination unit that determines whether the vehicle is parked at a location where people or vehicles often pass close to the vehicle, based on position information of the vehicle; and a recording function control unit that selectively performs a process of storing video data based on detection of a moving body by the moving body detection unit, and a process of storing video data based on acceleration detected by the acceleration detection unit, according to a determination result of the parking position determination unit.

Abstract: A system (1) for detecting dynamic secondary objects (55) that have a potential to intersect the trajectory (51) of a moving primary object (50), comprising a vision sensor (2) with a light-sensitive area (20) that comprises event-based pixels (21), so that a relative change in the light intensity impinging onto an event-based pixel (21) of the vision sensor (2) by at least a predetermined percentage causes the vision sensor (2) to emit an event (21a) associated with this event-based pixel (21), wherein the system (1) further comprises a discriminator module (3) that gets both the stream of events (21a) from the vision sensor (2) and information (52) about the heading and/or speed of the motion of the primary object (50) as inputs, and is configured to identify, from said stream of events (21a), based at least in part on said information (52), events (21b) that are likely to be caused by the motion of a secondary object (55), rather than by the motion of the primary object (50).

Abstract: A method for the automatic execution of a Static Dynamic Personality Test (TPSD) of a person, the method involving use of a cubic-spherical container obtained by merging concentrically a cube and a sphere. The method includes taking images of the person and overlapping the cubic-spherical container to the images, and measuring static-dynamic positions of the person as depicted in the images with respect to the cubic-spherical container.

Abstract: In implementations of resource-aware training for neural network, one or more computing devices of a system implement an architecture optimization module for monitoring parameter utilization while training a neural network. Dead neurons of the neural network are identified as having activation scales less than a threshold. Neurons with activation scales greater than or equal to the threshold are identified as survived neurons. The dead neurons are converted to reborn neurons by adding the dead neurons to layers of the neural network having the survived neurons. The reborn neurons are prevented from connecting to the survived neurons for training the reborn neurons.

Abstract: With the disclosed systems and methods for DRAM and 3D NAND inspection, an image of the wafer is received based on the output for an inspection tool. Geometric measurements of a design of a plurality of memory devices on the wafer are received. A care area with higher inspection sensitivity is determined based on the geometric measurements.

Abstract: The disclosure is related to a panoramic radiography device. The panoramic radiography device may include an image processor and a viewer module. The image processor may be configured to produce a primary panoramic image using a first image layer and a secondary panoramic image using a secondary image layer based on a plurality of image frame data, wherein the second image layer is different from the first image layer in at least one of a number, a position, a shape, an angle, and a thickness. The viewer module may be configured to i) provide a graphic user interface having a primary display area and a secondary display area arranged at a predetermined position of the primary display area, ii) display the primary panoramic image at the primary display area, and iii) display a part of the secondary panoramic image at the secondary display area, wherein the part of the secondary panoramic image corresponds to the predetermined position.

Abstract: Provided is a mobile apparatus obstacle detection system (100), a mobile apparatus (10) carrying the obstacle detection system (100), and a ground-sweeping robot. The obstacle detection system (100) comprises: a structured light projection module (102) configured to project structured light onto the path of advance of the mobile apparatus (10), the structured light comprising at least one lateral detection line in the horizontal direction and at least one longitudinal detection line in the vertical direction; a camera module (103) configured to capture an image of the structured light (105); and an image processing module (104) configured to calculate, according to the image of the structured light (105), the distances and positions of obstacles (106, 107, 108) on the path of advance.

Abstract: In one aspect, a method includes receiving first query fingerprint data representing first content channeled through a portion of a content-distribution system. The method also includes detecting a first match between the received first query fingerprint data and first reference fingerprint data representing a modifiable content-segment. The method also includes responsive to detecting the first match, performing a first action. The method also includes receiving second query fingerprint data representing content received by a content-presentation device. The method also includes detecting a second match between the received second query fingerprint data and second reference fingerprint data representing second content transmitted by the content-distribution system, where the second content is a modified version of the first content. The method also includes responsive to detecting the second match, performing a second action that is different from the first action.

Abstract: A method for aligning image frames, includes: obtaining a plurality of original image frames, the plurality of original image frames including a reference image frame and a plurality of to-be-aligned image frames; obtaining a brightness image of the reference image frame and a brightness image of each of the to-be-aligned image frames; for each of the to-be-aligned image frames, obtaining at least one mapping model between the reference image frame and the to-be-aligned image frame based on the brightness image of the reference image frame and the brightness image of the to-be-aligned image frame; and aligning the to-be-aligned image frame with the reference image frame based on the at least one mapping model.

Abstract: A method of transfer learning an object recognition neural network comprises acquiring a set of image frames; determining, by a first object recognition algorithm implementing an object recognition neural network, a plurality of object recognitions in the set of image frames; determining verified object recognitions by evaluating the plurality of object recognitions by a second, different from the first, object recognition algorithm, wherein an object recognition with a positive outcome in said evaluating forms a verified object recognition; forming a training set of annotated images comprising image frames associated with the verified object recognitions; performing transfer learning of the object recognition neural network based on the training set of annotated images.

Abstract: Embodiments of the disclosure provide systems and methods for detecting a medical condition of a subject. The system includes a communication interface configured to receive a sequence of images acquired from the subject by an image acquisition device and an end-to-end multi-task learning model. The end-to-end multi-task learning model includes an encoder, a Convolutional Recurrent Neural Network (ConvRNN), and at least one of a decoder and a classifier. The system further includes at least one processor configured to extract feature maps from the images using the encoder, capture contextual information between adjacent images in the sequence using the ConvRNN, and detect medical condition of the subject using the classifier based on the extracted feature maps of the image slices and the contextual information or segment each image slice using the decoder to obtain a region of interest indicative of the medical condition based on the extracted feature maps.

Abstract: A system and method include identification of a plurality of sub-volumes of an image volume, each of the plurality of sub-volumes of the image volume associated with a respective one of a plurality of vertebra, registration of each of the plurality of sub-volumes of the image volume to one of a plurality of reference sub-volumes associated with a same respective vertebra, input of the registered sub-volumes and the image volume to a trained neural network, reception of an output image volume from the trained neural network, the output image volume labeled to associate voxels with respective vertebra, and display of the output image volume.

Abstract: An improvement to automatic classifying of threat level of objects in CT scan images of container content, methods include automatic identification of non-classifiable threat level object images, and displaying on a display of an operator a de-cluttered image, to improve operator efficiency. The decluttered image includes, as subject images, the non-classifiable threat level object images. Improvement to resolution of non-classifiable threat objects includes computer-directed prompts for the operator to enter information regarding the subject image and, based on same, identifying the object type. Improvement to automatic classifying of threat levels includes incremental updating the classifying, using the determined object type and the threat level of the object type.

Abstract: There is provided a vehicle system including a sensing unit, a processing unit, a control unit and a display unit. The sensing unit is configured to capture an image frame containing an eyeball image from a predetermined distance. The processing unit is configured to calculate a pupil position of the eyeball image in the image frame and generate a drive signal corresponding to the pupil position. The control unit is configured to trigger a vehicle device associated with the pupil position according to the drive signal. The display unit is configured to show information of the vehicle device.

Abstract: A morphing part to be morphed and a morphing effect are acquired. Human body contour point information is acquired by detecting a human contour point of a human sideway image. A morphing region is determined according to the human contour point information and the morphing part. The morphing region is an image region to be morphed in the human sideway image. A human orientation is determined according to the human contour point information. A morphing direction of the morphing region is determined according to the human orientation and the morphing effect. Morphing is performed along the morphing direction in the morphing region.

Abstract: A system and method for recognizing objects in an image is described. The system can receive an image from a sensor and detect one or more objects in the image. The system can further detect one or more components of each detected object. Subsequently, the system can create a segmentation map based on the components detected for each detected object and determine whether the segmentation map matches a plurality of 3-D models (or projections thereof). Additionally, the system can display a notification through a user interface indicating whether the segmentation map matches at least one of the plurality of 3-D models.

Abstract: An object tracking system includes a sensor and a tracking system. The sensor is configured to capture a first frame of a global plane for at least a portion of a space. The tracking system is configured to receive a first coordinate in the global plane where a first marker is located in the space and to receive a second coordinate in the global plane where a second marker is located in the space. The tracking system is further configured to identify the first marker and the second marker within the first frame, to determine a first pixel location in the first frame for the first marker, to determine a second pixel location in the first frame for the second marker, and to generate a homography based on the first coordinate, the second coordinate, the first pixel location, and the second pixel location.