Abstract: An image processing device receives a caption for an image before, during, or after capture of the image by an image capture device. The image processing device generates image processing settings based on the caption, for instance based on a mood indicated in the caption or an object identified in the caption. If the caption is received before image capture, the image processing settings may include image capture settings that the image capture device may use to alter exposure or focus during image capture. Once the image is captured, the image processing device may process the image based on the image processing settings, for instance by applying filters or adjusting gain, brightness, contrast, saturation, or colors. For instance, brightness and saturation may be altered if the caption indicates a happy or sad mood, and focus may be altered to focus on an object identified in a caption.

Abstract: Disclosed herein are systems, devices, and processes for gesture detection. A method includes capturing a series of images. The method includes generating motion isolation information based on the series of images. The method includes generating a composite image based on the motion isolation information. The method includes determining a gesture based on the composite image. The processes described herein may include the use of convolutional neural networks on a series of time-related images to perform gesture detection on embedded systems or devices.

Abstract: The technology employs a patch-based multi-scale Transformer (300) that is usable with various imaging applications. This avoids constraints on image fixed input size and predicts the quality effectively on a native resolution image. A native resolution image (304) is transformed into a multi-scale representation (302), enabling the Transformer's self-attention mechanism to capture information on both fine-grained detailed patches and coarse-grained global patches. Spatial embedding (316) is employed to map patch positions to a fixed grid, in which patch locations at each scale are hashed to the same grid. A separate scale embedding (318) is employed to distinguish patches coming from different scales in the multiscale representation. Self-attention (508) is performed to create a final image representation. In some instances, prior to performing self-attention, the system may prepend a learnable classification token (322) to the set of input tokens.

Abstract: The present invention discloses a two-step anti-fraud vehicle insurance image collecting and quality testing method, system and device, the method comprises: step 1, collecting vehicle insurance scene images and marking vehicle orientation; step 2, performing object detection on the collected vehicle insurance scene images and screening to obtain object coordinates; step 3, according to the vehicle orientation and the object coordinates, obtaining the specific position of the object coordinates located in the whole vehicle; step 4, according to the object coordinates screened in step 2, performing vehicle component detection on the vehicle insurance scene images, obtaining the component coordinates of the vehicle components, and screening to obtain the vehicle component closest to the object coordinates; step 5, according to the specific position of the object coordinates located in the whole vehicle and the vehicle components closest to the object coordinates, obtaining the position of the vehicle components

Abstract: Segmenting an image is disclosed including acquiring a first image and a second image, the first image and the second image being obtained based on a same imaging target, and a resolution of the first image being greater than a resolution of the second image, performing image segmentation processing based on a plurality of sub-images of the first image to obtain a first initial segmentation result, performing image segmentation processing based on the second image to obtain a second initial segmentation result, merging, based on the imaging target, the first initial segmentation result with the second initial segmentation result to obtain a target segmentation result, and outputting or storing the target segment result.

Abstract: An image selection method, applied to a self-propelled apparatus, includes: collecting an image from a surrounding environment through an image collection device during the self-propelled apparatus travels; scoring the image according to a scoring rule when there is a recognizable obstacle in the collected image, wherein a value of the scoring is used to indicate an imaging quality of the recognizable obstacle in the image; and selecting an image that comprises the recognizable obstacle and that has a highest score as a to-be-displayed image in response to receive a request to view the image of the recognizable obstacle. A computer-readable storage medium and a self-propelled apparatus are further provided.

Abstract: The present disclosure provides an adaptive stereo matching optimization method, apparatus, and device, and a storage medium. The method includes: acquiring images of at least two perspectives of the same target scene, accordingly obtaining, through calculation, disparity value ranges corresponding to pixels in the target scene; and obtaining optimized depth value ranges by adjusting the disparity value ranges of the pixels in the target scene in real time through an adaptive stereo matching model; adjusting an execution cycle in the adaptive stereo matching model in real time through a DVFS algorithm according to a resource constraint condition of the processing system; and/or training on a plurality of scene image data sets through a convolutional neural network, so that the specific function parameters in the adaptive stereo matching model are correspondingly adjusted in real time according to the acquired different scene images.

Abstract: A method for identifying counterfeit coins, comprising receiving surface image data and edge image data of the coin at a processor. Identifying a plurality of defects using the processor. Comparing each of the plurality of defects to a database of known authentic coin image data defects to determine whether the coin is authentic.

Abstract: The disclosure relates to an artificial intelligence (AI) system using a machine learning algorithm such as deep learning and the like, and an application thereof. In particular, there is provided a control method for an electronic apparatus for searching for an image, the method comprising displaying an image comprising at least one object, detecting a user input for selecting an object, recognizing an object displayed at a point at which the user input is detected and acquiring information regarding the recognized object by using a recognition model trained to acquire information regarding an object, displaying a list including the information regarding the object, and based on one piece of information being selected from the information regarding the object included in the list, providing a related image by searching for the related image based on the selected information.

Abstract: Techniques and systems are provided for positioning mixed-reality devices within mixed-reality environments. The devices, which are configured to perform inside out tracking, transition between position tracking states in mixed-reality environments and utilize positional information from other inside out tracking devices that share the mixed-reality environments to identify/update positioning of the devices when they become disoriented within the environments and without requiring an extensive or full scan and comparison/matching of feature points that are detectable by the devices with mapped feature points of the maps associated with the mixed-reality environments. Such techniques can conserve processing and power consumption that would be required when performing a full or extensive scan and comparison of matching feature points. Such techniques can also enhance the accuracy and speed of positioning mixed-reality devices.

Abstract: Disclosed are methods, systems, and non-transitory computer-readable medium for analysis of images including wearable items. For example, a method may include obtaining a first set of images, each of the first set of images depicting a product; obtaining a first set of labels associated with the first set of images; training an image segmentation neural network based on the first set of images and the first set of labels; obtaining a second set of images, each of the second set of images depicting a known product; obtaining a second set of labels associated with the second set of images; training an image classification neural network based on the second set of images and the second set of labels; receiving a query image depicting a product that is not yet identified; and performing image segmentation of the query image and identifying the product in the image by performing image analysis.

Abstract: An annotation system uses annotations for a first set of sensor measurements from a first sensor to identify annotations for a second set of sensor measurements from a second sensor. The annotation system identifies reference annotations in the first set of sensor measurements that indicates a location of a characteristic object in the two-dimensional space. The annotation system determines a spatial region in the three-dimensional space of the second set of sensor measurements that corresponds to a portion of the scene represented in the annotation of the first set of sensor measurements. The annotation system determines annotations within the spatial region of the second set of sensor measurements that indicates a location of the characteristic object in the three-dimensional space.

Abstract: An exemplified methods and systems provides a Volterra filter network architecture that employs a cascaded implementation and a plurality of kernels, a set of which is configured to execute an nth order filter, wherein the plurality of kernels of the nth order filters are repeatedly configured in a plurality of cascading layers of interconnected kernels to form a cascading hierarchical structure that approximates a high-order filter.

Abstract: A manufacturing method of learning data is used for making a neural network perform learning. The manufacturing method of learning data includes a first acquiring step configured to acquire an original image, a second acquiring step configured to acquire a first image as a training image generated by adding blur to the original image, and a third acquiring step configured to acquire a second image as a ground truth image generated by adding blur to the original image. A blur amount added to the second image is smaller than that added to the first image.

Abstract: A method and apparatus for correcting an error in depth information estimated from a two-dimensional (2D) image are disclosed. The method includes diagnosing an error in depth information by inputting a color image and depth information estimated using the color image to a depth error detection network, and determining enhanced depth information by maintaining or correcting the depth information based on the diagnosed error.

Abstract: A system and methods for assessing road surface quality includes a wireless mobile device having a camera, a location receiver, and a road surface classifying computer application and is configured to be mounted on a vehicle. The system has a remote server having a road surface classifying web application, a database, and an interactive map connected to the web application. The mobile device actuates the camera to record videos, extract images from the videos, process the images, classify the images into road conditions, record a location of the images, generate a data packet including an identification of the mobile device and a time stamp of the data packet, and transmit the data packet to the remote server. The remote server stores the data packet in the database. The web application superimposes the time stamp of the data packet, the location, the road conditions, and the images on the interactive map.

Abstract: A method and system for reconstructing a magnetic particle distribution model based on time-frequency spectrum enhancement are provided. The method includes: scanning, by a magnetic particle imaging (MPI) device, a scan target to acquire a one-dimensional time-domain signal of the scan target; performing short-time Fourier transform to acquire a time-frequency spectrum; acquiring, by a deep neural network (DNN) fused with a self-attention mechanism, a denoised time-frequency spectrum; acquiring a high-quality magnetic particle time-domain signal; and reconstructing a magnetic particle distribution model. The method learns global and local information in the time-frequency spectrum through the DNN fused with the self-attention mechanism, thereby learning a relationship between different harmonics to distinguish between a particle signal and a noise signal.

Abstract: Examples of the present disclosure describe systems and methods for detecting and remediating compression artifacts in multimedia items. In example aspects, a machine learning model is trained on a dataset related to compression artifacts and non-compression artifacts. Input data may then be collected by a data collection module and provided to a pattern recognition module. The pattern recognition module may extract visual and audio features of the multimedia item and provide the extracted features to the trained machine learning model. The trained machine learning model may compare the extracted features to the model, and a confidence value may be generated. The confidence value may be compared to a confidence value threshold. If the confidence value is equal to or exceeds the confidence threshold, then the input data may be classified as containing at least one compression artifact. Remedial action may subsequently be applied (e.g., boosting the system with technical resources).

Abstract: The present application provides a method and a system for generating an image sample having a specific feature. The method includes: training a generative adversarial network-based sample generation model, where the generative adversarial network includes a generator and two discriminators: a global discriminator configured to perform global discrimination on an image, and a local discriminator configured to perform local discrimination on a specific feature; and inputting, to a trained generator that serves as a sample generation model, a semantic segmentation image that indicates a location of the specific feature and a corresponding real image not having the specific feature, to obtain a generated image sample having the specific feature.

Abstract: A person re-identification method based on a perspective-guided multi-adversarial attention is provided. The deep convolutional neural network includes a feature learning module, a multi-adversarial module, and a perspective-guided attention mechanism module. The multi-adversarial module is followed by a global pooling layer and a perspective discriminator after each stage of a basic network of the feature learning module. The perspective-guided attention mechanism module is an attention map generator and the perspective discriminator. The training of the deep convolutional neural network includes learning of the feature learning module, learning of the multi-adversarial module, and learning of the perspective-guided attention mechanism module. The proposed method uses the trained deep convolutional neural network to extract features of the testing images, and using an Euclidean distance to perform feature matching on images in a query set and images in a gallery set.