Abstract: A method performed by an electronic device is described. The method includes obtaining a first image from a first camera, the first camera having a first focal length and a first field of view. The method also includes obtaining a second image from a second camera, the second camera having a second focal length and a second field of view disposed within the first field of view. The method further includes aligning at least a portion of the first image and at least a portion of the second image to produce aligned images. The method additionally includes fusing the aligned images based on a diffusion kernel to produce a fused image. The diffusion kernel indicates a threshold level over a gray level range. The method also includes outputting the fused image. The method may be performed for each of a plurality of frames of a video feed.

Abstract: Disclosed herein is an image deep learning model training method. The method includes sampling a twin negative comprising a first negative sample and a second negative sample by selecting the first negative sample with a highest similarity out of an anchor sample and a positive sample constituting a matching pair in each class and by selecting the second negative sample with a highest similarity to the first negative sample, and training the samples to minimize a loss of a loss function in each class by utilizing the anchor sample, the positive sample, the first and second negative samples for each class. The first negative sample is selected in a different class from a class comprising the matching pair, and the second negative sample is selected in a different class from classes comprising the matching pair and the first negative sample.

Abstract: A method is presented. The method includes determining a number of landmarks in an image comprising multiple pixels. The method also includes determining a number of channels for the image based on a function of the number of landmarks. The method further includes determining, for each one of the number of channels, a confidence of each pixel of the multiple pixels corresponding to a landmark. The method still further includes identifying the landmark in the image based on the confidence.

Abstract: In a color printing environment, functions for printing color management are dissociated. An abstraction layer is also provided to facilitate setting and evaluation of all factors relating to color print and prediction.

Abstract: A method, apparatus, and system for generating tight two-dimensional (2D) bounding boxes for visible objects in a three-dimensional (3D) scene is disclosed. A two-dimensional (2D) segmentation image of a three-dimensional (3D) scene comprising one or more objects is generated by rendering the 3D scene with a segmentation camera. Each of the objects is rendered in a single respective different color. Next, one or more visible objects in the 3D scene are identified among the one or more objects based on the segmentation image. Next, a 2D amodal segmentation image for each of the visible objects in the 3D scene is generated separately. Each amodal segmentation image comprises only the single visible object for which it is generated. Thereafter, a 2D bounding box is generated for each of the visible objects in the 3D scene based on the amodal segmentation image for the visible object.

Abstract: Systems and methods are disclosed for training and using neural networks for computing depth maps. One method for training the neural network includes providing an image input to the neural network. The image input may include a camera image of a training scene. The method may also include providing a depth input to the neural network. The depth input may be based on a high-density depth map of the training scene and a sampling mask. The method may further include generating, using the neural network, a computed depth map of the training scene based on the image input and the depth input. The method may further include modifying the neural network based on an error between the computed depth map and the high-density depth map.

Abstract: Embodiments of the present disclosure provide a method and a device for detecting defect of a meal box, a server, a device, and a storage medium. The method includes: receiving a detection request including an image of the meal box sent by a user terminal, the image of the meal box being obtained by an image acquirer of the user terminal; and performing defect recognition based on the image of the meal box and a defect detection model in response to the detection request.

Abstract: A method and apparatus for image processing. A de-warping circuit is configured to retrieve, in an ordered sequence, a plurality of de-warping matrices for remapping pixels of a first image from a distorted image space to a corrected image space, where the first image depicts a scene in the distorted image space. The de-warping circuit reads, into a memory buffer, blocks of pixels of the first image based at least in part on the order of the de-warping matrices and generates a plurality of image tiles based on the blocks of pixels of the first image. Each of the image tiles is interpolated from one or more of the retrieved blocks using a respective one of the de-warping matrices. The de-warping circuit writes the plurality of image tiles to an external memory to produce a second image depicting the scene in the corrected image space.

Abstract: There is provided a projection instruction device that generates a projection image to be projected on parcel based on sensing information of the parcel, the device including: a processor; and a memory, in which by cooperating with the memory, the processor performs weighting on a value of a feature amount of a color image of parcel included in the sensing information based on a distance image of parcel included in the sensing information, and tracks the parcel based on the weighted value of the feature amount of the color image.

Abstract: Provided are a learning system, a method, and a program that can provide medical institutions with motivation for creating learning data related to medical images in a case of using machine learning to support diagnosis using medical images. The learning system includes a reception unit that receives an input of first learning data from a user, a calculation unit that calculates, a contribution degree of the first learning data to learning of a discriminator for each user on the basis of at least one of a comparison result between the first learning data and second learning data used for creating the discriminator or a comparison result between an output obtained by inputting the first learning data to the discriminator and correct answer data corresponding to the first learning data, and a service setting unit that sets a service for the user on the basis of the contribution degree calculated for each user.

Abstract: The purpose of the present invention is to provide a pattern matching device and computer program that carry out highly accurate positioning even if edge positions and numbers change. The present invention proposes a computer program and a pattern matching device wherein a plurality of edges included in first pattern data to be matched and a plurality of edges included in second pattern data to be matched with the first pattern data are associated, a plurality of different association combinations are prepared, the plurality of association combinations are evaluated using index values for the plurality of edges, and matching processing is carried out using the association combinations selected through the evaluation.

Abstract: Systems and methods are provided for receiving at least a first query color, and searching an electronic catalog including a plurality of product images for the first query color to determine a similarity measure between the first query color and a product image of a plurality of product images. The similarity measure may be determined by determining a Euclidean distance between values in a three-dimensional color space for the first query color and a target color of the product image, and determining the similarity measure between the query color and the product image by determining a sum of the similarity measures from all target colors on the product image, weighted by the coverage of each target color. The search results may be transmitted based on the searching of the electronic catalog including the plurality of product images for the first query color.

Abstract: The present disclosure provides a picture processing method and device, including: an integrated circuit chip IC receiving a to-be-processed picture sent by a graphics processor GPU; the IC pre-processing the to-be-processed picture; the IC performing counter-distortion process on the pre-processed picture; and the IC outputting the picture which is subjected to the counter-distortion process for display.

Abstract: A vehicular lighting system includes a light projecting device disposed at each headlight of a vehicle. The light projecting device is part of the respective headlight. The light projecting device, when operated, projects an animated sequence of visual images or a sequence of visual light patterns on a wall in front of the vehicle. The projected animated sequence of visual images or projected sequence of visual light patterns is viewable at the wall by a driver of the vehicle. The light projecting device may include a matrix of light emitting diodes. The visual images or visual light pattern projected by the light projecting device, when operated, may change as distance between the vehicle and the wall decreases.

Abstract: A machine accesses a set of image target models, each image target model being associated with model parameters, the model parameters comprising at least an operational domain, an expected input image quality, and an expected orientation. The machine receives an image for processing by one or more image target models from the set, the image including metadata specifying image parameters of the received image. The machine identifies, based on the image parameters in the metadata of the received image and the model parameters of one or more models in the set, a first subset of the set of image target models including image target models that are capable of processing the received image. The machine provides the received image to at least one image target model in the first subset.

Abstract: An object recognition device including an artificial neural network (NN) engine configured to receive learning data and weights, make an object recognition model (ORM) learn by using the received information, and provide selected weight data including weights from the selected portion of the weights, and further configured to receive a feature vector, and apply the feature vector extracted from an object data that constructs the object and the selected weight data to the learned ORM to provide an object recognition result, a nonvolatile memory (NVM) configured to store the learned ORM, and an error correction code (ECC) engine configured to perform an ECC encoding on the selected weight data to generate parity data, provide the selected weight data and the parity data to the NVM, and provide the selected weight data to the NN engine by performing an ECC decoding on the selected weight data based on the parity data.

Abstract: This disclosure relates to improved sketch-based image retrieval (SBIR) techniques. The SBIR techniques utilize a neural network architecture to train a domain migration function and a hashing function. The domain migration function is configured to transform sketches into synthetic images, and the hashing function is configured to generate hash codes from synthetic images and authentic images in a manner that preserves semantic consistency across the sketch and image domains. The hash codes generated from the synthetic images can be used for accurately identifying and retrieving authentic images corresponding to sketch queries, or vice versa.

Abstract: The present invention discloses a method and device for automatically drawing structural cracks and precisely measuring widths thereof. The method comprises a method for automatically drawing cracks and a method for calculating widths of these cracks based on a single-pixel skeleton and Zernike orthogonal moments, wherein the method for automatically drawing cracks is used to rapidly and precisely draw cracks in the surface of a structure, and the method for calculating widths of these cracks based on a single-pixel skeleton and Zernike orthogonal moments is used to calculate widths of macro-cracks and micro-cracks in an image in a real-time manner.

Abstract: Systems and methods of script identification in audio data obtained from audio data. The audio data is segmented into a plurality of utterances. A script model representative of a script text is obtained. The plurality of utterances are decoded with the script model. A determination is made if the script text occurred in the audio data.

Abstract: An automatic image annotation system receives a reference image with one or more parts annotated along with one or more query images and automatically identifies portions from the query images that are similar to the annotated parts of the reference image. The S-matrices of the reference image and the query images are obtained via singular value decomposition (SVD). Lower-dimensional images are also obtained for the reference image and the query images using a pre-trained deep learning model. The S-matrices and the lower-dimensional images of the corresponding images are combined to generate vector representations. A distance metric is calculated for the vector representation of the reference image with that of the query image. A preliminary output image with a preliminary annotation is initially generated. The preliminary annotation is further optimized to generate an optimized annotation that adequately covers the region of interest (ROI).