Abstract: An electronic device and method for three-dimensional (3D) reconstruction based on camera motion information is provided. The electronic device receives a set of images of a three-dimensional (3D) physical space captured by one or more image sensors. The electronic device further receives metadata associated with each of the set of images. The metadata may include at least motion information associated with the one or more image sensors that captured the set of images. The electronic device applies a neural network model on the received metadata. The electronic device determines a first set of images from the received set of images based on the application of the neural network model on the received metadata. The electronic device constructs a 3D model of a subject associated with the 3D physical space based on the determined first set of images.

Abstract: A method and system for synchronization of image data is provided. A plurality of image-capture devices is controlled to acquire a plurality of video clips of at least one object. Each video clip of the plurality of video clips is acquired by a corresponding image-capture device of the plurality of image-capture devices in a moving state. From the plurality of image-capture devices, a set of sensor data is acquired. Such data corresponds to the plurality of image-capture devices and is associated with a movement of the plurality of image-capture devices. Thereafter, relative offsets between the set of sensor data are determined by using cross-correlation and matching frames in each of the plurality of video clips is further determined, based on the relative offsets.

Abstract: A system and method for recognition of license plate numbers from Bayer-domain image data is provided. The system acquires Bayer-domain image data of a scene comprising a vehicle and inputs such image data to a Deep Neural Network (DNN) model trained for license plate detection (LPD) and license plate recognition (LPR). The system extracts an LPD result as an output of the DNN model. Based on the LPD result, the system extracts, from the input Bayer-domain image data, a region-of-interest (RoI) comprising a license plate image associated with the vehicle. The system inputs the extracted RoI to the DNN model and extracts, as an output of the DNN model, an LPR result which indicates a confidence of the DNN model in recognition of a license plate number of the vehicle from the input RoI. The system determines the license plate number based on the extracted LPR result.

Abstract: A system and method for recognition of license plate numbers from Bayer-domain image data is provided. The system acquires Bayer-domain image data of a scene comprising a vehicle and inputs such image data to a Deep Neural Network (DNN) model trained for license plate detection (LPD) and license plate recognition (LPR). The system extracts an LPD result as an output of the DNN model. Based on the LPD result, the system extracts, from the input Bayer-domain image data, a region-of-interest (RoI) comprising a license plate image associated with the vehicle. The system inputs the extracted RoI to the DNN model and extracts, as an output of the DNN model, an LPR result which indicates a confidence of the DNN model in recognition of a license plate number of the vehicle from the input RoI. The system determines the license plate number based on the extracted LPR result.

Abstract: An electronic device and method for three-dimensional (3D) reconstruction based on camera motion information is provided. The electronic device receives a set of images of a three-dimensional (3D) physical space captured by one or more image sensors. The electronic device further receives metadata associated with each of the set of images. The metadata may include at least motion information associated with the one or more image sensors that captured the set of images. The electronic device applies a neural network model on the received metadata. The electronic device determines a first set of images from the received set of images based on the application of the neural network model on the received metadata. The electronic device constructs a 3D model of a subject associated with the 3D physical space based on the determined first set of images.

Abstract: A method and system for synchronization of image data is provided. A plurality of image-capture devices is controlled to acquire a plurality of video clips of at least one object. Each video clip of the plurality of video clips is acquired by a corresponding image-capture device of the plurality of image-capture devices in a moving state. From the plurality of image-capture devices, a set of sensor data is acquired. Such data corresponds to the plurality of image-capture devices and is associated with a movement of the plurality of image-capture devices. Thereafter, relative offsets between the set of sensor data are determined by using cross-correlation and matching frames in each of the plurality of video clips is further determined, based on the relative offsets.

Abstract: An electronic apparatus and method for traffic speed enforcement based on license plate detection is provided. The electronic apparatus controls an image-capture device on a first vehicle to capture a sequence of images of a second vehicle and detects a license plate of the second vehicle in the captured sequence of images. The electronic apparatus extracts a set of license plate images from a set of images in the sequence of images and determines a first change in a size of a first license plate image with respect to that of a second license plate image. The electronic apparatus further determines a speed of the first vehicle and a traffic speed violation by the second vehicle based on the determined first change and the determined speed of the first vehicle. The electronic apparatus further controls a media device to output a notification indicative of the determined traffic speed violation.

Abstract: An electronic apparatus and method for traffic speed enforcement based on license plate detection is provided. The electronic apparatus controls an image-capture device on a first vehicle to capture a sequence of images of a second vehicle and detects a license plate of the second vehicle in the captured sequence of images. The electronic apparatus extracts a set of license plate images from a set of images in the sequence of images and determines a first change in a size of a first license plate image with respect to that of a second license plate image. The electronic apparatus further determines a speed of the first vehicle and a traffic speed violation by the second vehicle based on the determined first change and the determined speed of the first vehicle. The electronic apparatus further controls a media device to output a notification indicative of the determined traffic speed violation.

Abstract: Before a start frame from which decoding of a bit stream is to be started and a reference image frame which is decoded in order to obtain a reference image necessary to decode the start frame are decoded, a quantization matrix stored in a quantization matrix buffer is updated such that, where a quantization matrix is described in any of B pictures disposed between the start frame to be decoded or the reference image frame and an I picture or a P picture preceding in time to the start frame or the reference image frame, the start frame and the reference image frame are decoded with the quantization matrix described in the B picture. Then, where a quantization matrix is described in the frame to be decoded, a decoding process of the bit stream is performed using the quantization matrix and the quantization matrix buffer is updated using the quantization matrix.

Abstract: Disclosed are systems and methods for storage device access control initiated by a non-linear editor (NLE). In one embodiment, a storage device access controller can include: a storage device configured to store clips, such as audio, graphics, or video clips, arranged in frames; an NLE coupled to the storage device for requesting one or more of the clips; and a buffer engine coupled to the storage device and the NLE. The buffer engine can include: buffers for storing data from the storage device; a data index for indicating a location of data in a clip on the storage device; and a reader configured to control an access of the data using the data index in response to one or more parameters. Embodiments of the present invention can provide for improved data access performance from a disk where the data is requested in clip form from an NLE.