Abstract: An apparatus includes an extraction unit configured to extract an N-dimensional feature vector (N is an integer that is greater than M) in a second format including an M-dimensional feature vector (M is an integer of 2 or greater) in a first format from input data, and an identification unit configured to identify a target in the input data based on the feature vector in the first format and the feature vector in the second format.

Abstract: An image processing apparatus that processes images acquired by a camera. In processing the images, a moving object may be recognized or identified from the images. Masking processing is performed on the moving object recognized in the image. An estimate of the spatial motion of the image processing apparatus relative to a surrounding environment is determined based on the images that are subjected to the masking processing.

Abstract: Provided are a multimodal image matching method and system, terminal device, and storage medium. The method includes: performing self-supervised feature extraction on an optical image and a synthetic aperture radar (SAR) image to obtain a repetitive feature point between the optical image and the SAR image; segmenting the optical image and the SAR image into a first image block sequence based on the repetitive feature point, and performing feature extraction on the first image block sequence through a dual-branch network to obtain feature description vectors of the optical image and the SAR image respectively, where the dual-branch network includes a first branch network for extracting a global feature and a second branch network for extracting a local feature; and performing feature matching on the optical image and the SAR image based on the feature description vectors to obtain a matching point pair between the optical image and the SAR image.

Abstract: An information processing device is a device used in a mounting device that collects a component and arranges the component on a substrate. In this information processing device, a control section acquires a captured image including a component, sets multiple detection lines for detecting a brightness difference of the component with respect to the acquired captured image in the component, and obtains a reference position of an outer edge portion from the multiple detection lines. Next, the control section uses a predetermined weight coefficient of which weight tends to decrease as a distance from the reference position increases to obtain a candidate value obtained by adding the weight coefficient to one or more outer edge portion candidates existing on the detection line. Subsequently, the control section selects a position of the outer edge portion existing on the detection line based on the obtained candidate value.

Abstract: In some implementations, a robot host may receive a video associated with assembly using a plurality of sub-objects. The robot host may determine spatio-temporal features based on the video and may identify a plurality of actions represented in the video based on the spatio-temporal features. The robot host may map the plurality of actions to the plurality of sub-objects to generate an assembly plan and may combine output from a point cloud model and output from a color embedding model to generate a plurality of sets of coordinates corresponding to the plurality of sub-objects. The robot host may perform object segmentation to estimate a plurality of grip points and a plurality of widths corresponding to the plurality of sub-objects. Accordingly, the robot host may generate instructions, for robotic machines, based on the assembly plan, the plurality of sets of coordinates, the plurality of grip points, and the plurality of widths.

Abstract: An indication system for a surgical lighting apparatus is provided, comprising a computer device connectable to a plurality of light modules, wherein the computer device is configured to determine an illumination setting for each of the plurality of light modules; compare the illumination settings of at least two of the plurality of light modules to at least one output illumination threshold; and output an indication if the at least one output illumination threshold is exceeded. A surgical lighting apparatus, a method of indication, and a detection system for a surgical lighting apparatus are also provided.

Abstract: An image processing apparatus is configured to determine presence or absence of print abnormality in a target image by comparing a reference image or reference images, each of which is a print result to be a reference, and the target image, which is a print result to be an inspection target. The image processing apparatus includes a setting unit configured to set an inspection condition for a printed product, and a comparison unit configured to compare the reference image or reference images and the target image using a number of the reference images set based on the inspection condition.

Abstract: A method of depth map completion is described. A color map and a sparse depth map of a target scenario can be received. Resolutions of the color map and the sparse depth map are adjusted to generate n pairs of color maps and sparse depth maps of n different resolutions. The n pairs of color maps and the sparse depth maps can be processed to generate n prediction result maps using a cascade hourglass network including n levels of hourglass networks. Each of the n pair is input to a respective one of the n levels to generate the respective one of the n prediction result maps. The n prediction result maps each include a dense depth map of the same resolution as the corresponding pair. A final dense depth map of the target scenario can be generated according to the dense depth maps.

Abstract: Embodiments described herein provide systems, methods, and devices for generating enhanced vison-language training data. A method may include: receiving, from a communication interface, a first training dataset of image-text pairs and a second training dataset of annotated image-text pairs; fine-tuning an image-grounded text decoder and an image-grounded text encoder using the second training dataset of annotated image-text pairs; generating, by the fine-tuned image-grounded text decoder, a predicted text based on a training image from the first training dataset; generating, by the fine-tuned image-grounded text encoder, a filtering decision based on the training image and the predicted text; adding the training image and the predicted text to form a third training dataset of image-text pairs depending on the filter decision; and training a vision-language model using the third training dataset of image-text pairs.

Abstract: An information processing apparatus includes a movable object detector that detects a movable object from a captured image captured with a fisheye camera, a human determiner that determines whether the movable object is a human by comparing a distance between two predetermined points on an outline of a movable object area including the movable object with a threshold range set based on a height of the human measured at a position of the movable object in the captured image, and a human detector that detects the human from the movable object area including the movable object determined as the human by the human determiner.

Abstract: An information processing apparatus, for estimating a movement amount of a moving object to which the information processing apparatus is attached, captures an image taken of an exterior view seen from the moving object. The apparatus estimates the movement amount of the moving object based on a first taken image taken at a first time and a second taken image taken at a second time earlier than the first time. The apparatus estimates the movement amount of the moving object based on a difference between image information on a predetermined region of each manipulated image obtained by manipulating the second taken image based on mutually different predicted movement amounts and image information on a predetermined region cut out from the first taken image.

Abstract: Systems and methods for detecting crop marks depicted in digital images are disclosed. According to certain aspects, an electronic device detects a set of crop marks in a reference digital image and aligns the reference digital image with an input digital image. Based on locations of the detected crop marks as aligned to the input digital image, the electronic device detects a set of partial crop marks in the input digital image. As a result, the partial crop marks in the input digital image may be removed.

Abstract: A video coding mechanism is disclosed. The mechanism includes receiving a bitstream comprising a plurality of sub-pictures partitioned from a picture such that a union of the sub-pictures covers a total area of the picture without overlap. The bitstream is parsed to obtain the one or more sub-pictures. The one or more sub-pictures are decoded to create a video sequence. The video sequence is forwarded for display.

Abstract: A method for identifying inaccurately depicted boxes in an image, such as miss detected boxes and partially detected boxes. The method obtains a 2D RGB image of the boxes and a 2D depth map image of the boxes using a 3D camera, where pixels in the depth map image are assigned a value identifying the distance from the camera to the boxes. The method generates a segmentation image of the boxes using a neural network by performing an image segmentation process that extracts features from the RGB image and segments the boxes by assigning a label to pixels in the RGB image so that each box in the segmentation image has the same label and different boxes in the segmentation image have different labels. The method analyzes the segmentation image to determine if the image segmentation process has failed to accurately segment the boxes in the segmentation image.

Abstract: A system comprising a first image processing arrangement and a second image processing arrangement, wherein the first image processing arrangement comprises a controller configured to: a) receive an image; b) select a task and a task identifier associated with task data; c) compress the image based on the task data; and d) transmit the compressed image to the second image processing arrangement for processing, and wherein the second image processing arrangement comprises a controller configured to: e) receive the compressed image and task identifier; f) retrieve task parameters associated with the task identifier; g) process the compressed image based on the task parameters; h) determine results and i) transmit the at least indications of the determined results to the first image processing arrangement, and the controller of the first image processing arrangement is further configured to: j) receive at least indications of a result of the processing from the second image processing arrangement; and k) indicat

Abstract: A method of providing tactical information for team sports is disclosed. The method includes acquiring a plurality of player tracking data sets for a plurality of players, acquiring role assignment information using the plurality of player tracking data sets, determining a dominant role assignment based on the plurality of role assignments, determining at least one irregular role assignment from the plurality of role assignments in consideration of the dominant role assignment, acquiring a plurality of space information sets associated with the target period for each of the plurality of role indices based on the remaining role assignment information other than the at least one irregular role assignment and some of the player tracking data sets corresponding to the remaining role assignment information and updating the role assignment information using the plurality of space information sets.

Abstract: A method for processing a video includes determining characteristics regarding a current video block; determining an operational state regarding unsymmetrical quad-tree (UQT) splitting based on the determination of the characteristics, wherein the operational state indicates that whether the UQT splitting is enabled or disabled and/or indicates how to split the current video block into four partitions using the UQT splitting; and performing a conversion between the current video block and a bitstream representation of the current video block based on the operational state of UQT splitting.

Abstract: Disclosed herein are image encoding/decoding method and apparatus. The image decoding method according to the present invention includes: obtaining prediction mode information of a neighboring block for a current block from a bitstream; by using the obtained information, deriving a prediction mode of the neighboring block; and, based on the prediction mode of the neighboring block, reconstructing the current block.

Abstract: Surgical kit inspection systems and methods are provided for inspecting surgical kits having parts of different types. The surgical kit inspection system comprises a vision unit including a first camera unit and a second camera unit to capture images of parts of a first type and a second type in each kit and to capture images of loose parts from each kit that are placed on a light surface. A robot supports the vision unit to move the first and second camera units relative to the parts in each surgical kit. One or more controllers obtain unique inspection instructions for each of the surgical kits to control inspection of each of the surgical kits and control movement of the robot and the vision unit accordingly to provide output indicating inspection results for each of the surgical kits.

Abstract: Disclosed are computer systems and computer-implemented methods for training a machine-learning method for determining pre-determined points in an image. The method includes training a first machine-learning method based on training data indicating whether or not a pre-determined point is present in a cell of the image; training a second machine-learning method based on a position of pre-determined points in the cell and based on training of the first machine-learning method; and training a third machine-learning method based on a modified version of the image and based on training of the second machine-learning method. Also disclosed are methods and computer systems implementing the trained machine-learning method.