Abstract: In various examples, apparatuses, systems, and techniques to perform offline image signal processing of source image data to generate target image data. In at least one embodiment, data collection using exposure and calibration setting of an image sensor is performed to generate source image data, which is then processed by using offline image signal processing to generate target data.

Abstract: A data analysis system to analyze data. The data analysis system includes a data buffer configured to receive data to be analyzed. The data analysis system also includes a state machine lattice. The state machine lattice includes multiple data analysis elements and each data analysis element includes multiple memory cells configured to analyze at least a portion of the data and to output a result of the analysis. The data analysis system includes a buffer interface configured to receive the data from the data buffer and to provide the data to the state machine lattice.

Abstract: A method for simulating sharpness of an image boundary including setting a boundary pattern including a boundary line between black and white, and a unit pattern including arrangement information of pixels; calculating simulation grayscale data for implementing the boundary pattern by using the unit pattern; and obtaining a sharpness index for the boundary line based on a contrast sensitivity function (CSF) and brightness data extracted from the simulation grayscale data.

Abstract: A system for low compute high-resolution depth map generation using low-resolution cameras is configured to obtain a stereo pair of images and generate a depth map by performing stereo matching on the stereo pair of images. The system is also configured to obtain a first image comprising first texture information for the environment that has a first image resolution that is higher than an image resolution of images of the stereo pair of images. The system is further configured to generate a reprojected first image by reprojecting the first image to correspond to an image capture perspective associated with the depth map. The reprojection of the first image is based on depth information from the depth map and includes reprojected first texture information for the environment. The system is also configured to generate an upsampled depth map based on the depth map.

Abstract: Systems, apparatuses, and methods for performing parallel histogram calculation with application to palette table derivation are disclosed. An encoder calculates a first histogram for a first portion of pixel component value bits of a block of pixels. Then, the encoder selects a first number of the highest pixel count bins from the first histogram. Also, the encoder calculates a second histogram for a second portion of pixel component value bits of the block. The encoder selects a second number of the highest pixel count bins from the second histogram. A third histogram is calculated from the concatenation of bits assigned to the first and second number of bins, and the highest pixel count bins are selected from the third histogram. A palette table is derived based on these highest pixel count bins selected from the third histogram, and the block of pixels is encoded using the palette table.

Abstract: An information processing device includes a processing unit configured to divide original image data for a larger number of channels Nb than the number of channels Na into a plurality of groups so that a number of channels of each of the groups becomes no larger than the number of channels Na, generate compressed image data for the Na channels according to the processing unit from the original image data included in each of the groups, and generate the print command including the compressed image data for the Na channels corresponding to each of the groups, and a communication unit configured to transmit the print command to the printer.

Abstract: A system for producing a textile component of an article includes an additive manufacturing device in selective communication with a processor and memory. The processor and memory are configured to determine a strain value in a region of the textile component of the article based on images of the article from a camera in selective communication with the processor and memory and to generate a strain map based on the strain value. The additive manufacturing device is configured to apply a reinforcement to a textile substrate to variably reinforce the textile substrate according to the strain map and to form the textile component of the article.

Abstract: An apparatus comprising means for: receiving values for sub-bands of a frame of an audio signal, the values comprising at least one azimuth value, at least one elevation value at least one energy ratio value and at least one spread and/or surround coherence value for each sub-band; determining a codebook for encoding at least one spread and/or surround coherence value for each sub-band based on the at least one energy ratio value and at least one azimuth value for each sub-band for a frame; discrete cosine transforming at least one vector, the at least one vector comprising the at least one spread and/or surround coherence value for a sub-band for the frame; and encoding a first number of components of the discrete cosine transformed vector based on the determined codebook.

Abstract: The disclosure regards cross-component prediction and methods for deriving of a linear model for obtaining a first-component sample for a first-component block from an associated reconstructed second-component sample of a second-component block in the same frame, the method comprising determining the parameters of a linear equation representing a straight line passing through two points, each point being defined by two variables, the first variable corresponding to a second-component sample value, the second variable corresponding to a first-component sample value, based on reconstructed samples of both the first-component and the second-component; and deriving the linear model defined by the straight line parameters; wherein said determining the parameters uses integer arithmetic.

Abstract: An apparatus to facilitate compression of memory data is disclosed. The apparatus comprises one or more processors to receive uncompressed data, adapt a format of the uncompressed data to a compression format, perform a color transformation from a first color space to a second color space, perform a residual computation to generate residual data, compress the residual data via entropy encoding to generate compressed data and packing the compressed data.

Abstract: Methods and apparatus for parsing friendly and error resilient merge flag coding in video coding are provided. In some methods, in contrast to merging candidate list size dependent coding of the merge flag in the prior art, a merge flag is always encoded in the encoded bit stream for each inter-predicted prediction unit (PU) that is not encoded using skip mode. In some methods, in contrast to the prior art that allowed the merging candidate list to be empty, one or more zero motion vector merging candidates formatted according to the prediction type of the slice containing a PU are added to the merging candidate list if needed to ensure that the list is not empty and/or to ensure that the list contains a maximum number of merging candidates.

Abstract: This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for GPU wave-to-wave optimization. A graphics processor may execute a shader program for a first wave associated with a draw call or a compute kernel. The graphics processor may identify at least one first indication for the first wave associated with the draw call or the compute kernel. The graphics processor may store the at least one first indication for the first wave to a memory location. The graphics processor may execute the shader program for at least one second wave associated with the draw call or the compute kernel. The execution of the shader program for the at least one second wave may be based on the shader program for the at least one second wave reading the memory location to retrieve the at least one first indication.

Abstract: Methods and apparatus for video processing are described. The video processing may include video encoding, video decoding or video transcoding. One example video processing method includes performing a conversion between a video including one or more pictures including one or more slices and a coded representation of the video, wherein the coded representation conforms to a format rule that specifies that a condition of a field in the coded representation controls a constraint on a slice type or whether the slice type is included in the coded representation for a video slice, wherein the field includes a general constraint flag, a network abstraction layer unit type or whether the video slice is in a first video picture of an access unit.

Abstract: Decompressing a compressed image to obtain a decompressed image includes receiving, in a compressed stream, compressed pixel values of the compressed image; decompressing, from the compressed stream, a first compressed pixel value of the compressed pixel values using a lossy floating-point decompression scheme to obtain a floating-point pixel value; rounding the floating-point pixel value to a nearest integer to obtain a pixel value of the decompressed image; and displaying or storing the decompressed image.

Abstract: A data analysis system to analyze data. The data analysis system includes a data buffer configured to receive data to be analyzed. The data analysis system also includes a state machine lattice. The state machine lattice includes multiple data analysis elements and each data analysis element includes multiple memory cells configured to analyze at least a portion of the data and to output a result of the analysis. The data analysis system includes a buffer interface configured to receive the data from the data buffer and to provide the data to the state machine lattice.

Abstract: A display apparatus and a controlling method thereof are disclosed. The display apparatus includes: a memory storing one or more instructions; and a processor configured to execute the stored one or more instructions to: obtain encoded data of a first digital image and artificial intelligence (AI) meta-information indicating a specification of a deep neural network (DNN), obtain a second digital image corresponding to the first digital image by decoding the encoded data, obtain a light signal converted from the second digital image according to a previously determined electro-optical transfer function (EOTF), and obtain a display signal by processing the light signal by using an opto-optical transfer function (OOTF) and a high dynamic range (HDR) DNN set according to the AI meta-information.

Abstract: It is provided a recognition device for analyzing a starvation extent of a whiteleg shrimp based on underwater imaging, including a bracket, a camera mounted on a top of the bracket, a plurality of light sources for illumination, and a processor connected with the camera. The processor receives a shrimp image collected by the camera, extracts an edge image of the shrimp after the collected shrimp image is preprocessed, and calculates a movement speed of the shrimp and a quantity of baits after a head and a tail are recognized, to recognize starvation extent of the shrimp. It is also provided a recognition method for analyzing starvation extent of the whiteleg shrimp based on underwater imaging. The present disclosure may guide feeders to feed and realize reasonable culture for the shrimp, by capturing images of the shrimp in time and determining starvation the extent of shrimp.

Abstract: A display device includes a display panel and a driver which receives image signals and transmits data signals to the display panel. The driver includes an image sticking compensator that converts the image signals such that the first image is periodically shifted while being displayed. The image sticking compensator includes an extractor which extracts compensation area data corresponding to a first image displayed in a compensation area, a calculator which calculates fixed data based on the compensation area data and corresponding to the first image, and a shifter which generates shift-fixed data based on the fixed data. The compensation area includes a first area in which the first image is displayed and a second area in which a peripheral image at least partially surrounding the first image is displayed.

Abstract: The disclosure regards cross-component prediction and methods for deriving of a linear model for obtaining a first-component sample for a first-component block from an associated reconstructed second-component sample of a second-component block in the same frame, the method comprising determining the parameters of a linear equation representing a straight line passing through two points, each point being defined by two variables, the first variable corresponding to a second-component sample value, the second variable corresponding to a first-component sample value, based on reconstructed samples of both the first-component and the second-component; and deriving the linear model defined by the straight line parameters; wherein said determining the parameters uses integer arithmetic.

Abstract: A method of constraining data represented in a deep neural network is described. The method includes determining an initial shifting specified to convert a fixed-point input value to a floating-point output value. The method also includes determining an additional shifting specified to constrain a dynamic range during converting of the fixed-point input value to the floating-point output value. The method further includes performing both the initial shifting and the additional shifting together to form a dynamic, range constrained, normalized floating-point output value.

Abstract: A distributed event processing system is disclosed that receives a batch of events via a continuous data stream and performs the serialization of data in the batch of events. In certain embodiments, the system identifies a first data type of a first attribute for each event in a batch of events and determines a first type of data compression to be performed on data values represented by the first attribute. The system determines a first type of data compression to be performed on data values represented by the first attribute based on the first data type of the first attribute. The system then generates a first set of serialized data values for the first attribute. The system processes the first set of serialized data values against a set of one or more continuous queries to generate a first set of output events.

Abstract: An image sensor and an image processing system in which the image sensor includes a sensing unit configured to generate a plurality of images having different luminances with respect to a same object, a pre-processor configured to merge n images (n is a natural number equal to or greater than 2) except for at least one of the plurality of images to generate a merged image, and an interface circuit configured to output the at least one image and the merged image to an external processor.

Abstract: An encoding method is provided which includes receiving a plurality of images, obtaining values of elements in a portion of the images, sorting the elements according to different values of the elements, sorting the elements according to a number of occurrences of the different values and encoding the elements using a subset of the different values having corresponding numbers of occurrences that are higher than corresponding numbers of occurrences of other values. Examples also include a processing device and method for use with palette mode encoding in which the elements are a portion of pixels in images and the values are color values of the portion of pixels in the images.

Abstract: A method for image-processing and an electronic device are disclosed. The method includes: obtaining a RAW image packet comprising at least two RAW images, the at least two RAW images having different exposure durations; unpacking the RAW image packet and obtaining the at least two of RAW images; obtaining a High Dynamic Range (HDR) RAW image by performing an image synthesis operation on the at least two RAW images; and performing a previewing, photographing, or video-recording operation on the HDR RAW image.

Abstract: An image processing apparatus including at least one processor coupled to a memory, serving as a first obtainment unit to obtain a luminance value of image data corresponding to a first luminance range, a second obtainment unit to obtain information for a second luminance range that is less than the first luminance range, a classification unit to classify, based on a correspondence relationship of luminance value conversion from the first luminance range to the second luminance range, the first luminance range of the image data into a plurality of regions, and a display unit to cause a display device to display an image based on the image data such that a pixel of the image having a luminance value belonging to a region, in which a tone characteristic in the image data can be restored when the luminance value conversion is performed, can be specified in the displayed image.

Abstract: The invention discloses a point cloud geometric compression method based on a depth auto-encoder, which comprises the following steps of: (1) point cloud preprocessing: collecting a type of point cloud to be compressed as a training set, and normalizing the training set into a unit circle; (2) point cloud down-sampling: down-sampling the point clouds in the training set so that each point cloud have the same point number m and each point in the point cloud has (x, y, z) three-dimensional coordinates; and adopting a farthest point down-sampling method by randomly selecting a point for the first time and then selecting the point farthest away from the selected point set every time to add into the selected point set until the selected point number meets the requirement; (3) training a compression model: inputting the point cloud sampled in step (2) into a point cloud geometric compression framework based on a depth auto-encoder for training; and (4) geometric compression of the point cloud: applying the trained

Abstract: A data compression apparatus includes: a compression unit that performs a compression process on output data that is sequentially outputted as time series data by a monitoring apparatus for monitoring a monitoring target range; an output unit that outputs the output data on which the compression process is performed, to a data processing apparatus that performs an object detection process for detecting an object that exists in the monitoring target range by using the output data; and a setting unit that sets a compression ratio used in the compression unit, on the basis of an accuracy information indicating a relationship between a compression ratio at which the output data is compressed and a detection accuracy of the object by the object detection process performed by using the output data compressed at the compression ratio.

Abstract: An apparatus configured for head-worn by a user, includes: a screen configured to present graphics for the user; a camera system configured to view an environment in which the user is located; and a processing unit coupled to the camera system, the processing unit configured to: obtain a first image with a first resolution, the first image having a first corner, determine a second image with a second resolution, the second image having a second corner that corresponds with the first corner in the first image, wherein the second image is based on the first image, the second resolution being less than the first resolution, detect the second corner in the second image, determine a position of the second corner in the second image, and determine a position of the first corner in the first image based at least in part on the determined position of the second corner in the second image.

Abstract: A classifier training method includes detecting error data from training data; and training a classifier configured to detect an object based on the error data.

Abstract: This application provides a method for updating a route in a network. The first network device sends a first LSA packet to a third network device, so that the third network device generates a first route whose destination address is a second IP address, where a next-hop IP address of the first route is the IP address of the first network device, and the second IP address belongs to the IP network segment corresponding to the first IP address which is an IP address of the first network device. The first network device sends a second LSA packet to the third network device when determining that switching needs to be performed on a next hop of a route in the third network device, whose destination address belongs to the IP network segment, and whose next-hop IP address is the first IP address.

Abstract: The disclosure relates to the transfer of per-pixel transparency information using video codecs that do not provide an alpha channel (alternatively referred to as “transparency-agnostic video codecs”). For example, alpha information of visual elements may be transcoded into the supported channels of a video stream to generate additional samples of a supported color space, which are representative of the alpha information. After being encoded by a “transparency-agnostic video codec” and transmitted, the received alpha information may then be extracted from the supported channels of the video stream to render the received visuals with corresponding per-pixel transparency.

Abstract: Included are: an image acquiring unit acquiring a determination target image in which a target object to be inspected is captured; a learning result acquiring unit acquiring a result of machine learning of forward domain conversion of an image or inverse domain conversion of an image performed using a normal image in which the target object in a normal state is captured as training data; a determination target image analysis unit obtaining a domain-converted image by sequentially performing forward domain conversion and inverse domain conversion on the determination target image using the result of the machine learning; and a determination unit determining whether or not an abnormality is occurring in the target object captured in the determination target image by comparing the determination target image and the domain-converted image.

Abstract: The subject technology provides a messaging application in which an image can be captured and supplemented with supplemental content such as stickers, animations, etc., from within an active messaging thread. In this manner, a user participating in a messaging conversation with a remote user can add stickers, animations, and/or adaptive content to an image captured by the user, without having to locally cache/store the captured image before editing and without having to leave the messaging conversion (or the messaging application) to access an image editing application.

Abstract: An electronic apparatus includes: a storage and a processor. The processor is configured to: select a second pixel having a similarity to a first pixel equal to or greater than a threshold value among neighboring pixels based on a first chrominance signal of the first pixel and second chrominance signals of pixels neighboring the first pixel among a plurality of pixels included in an input image, identify index information corresponding to a pixel combination having a highest similarity to the first pixel based on a first luminance signal of the first pixel, a second luminance signal of the second pixel, and a plurality of pixel combinations with respect to the first pixel, and control the electronic apparatus to store the image by replacing a pixel value of the first pixel with the identified index information.

Abstract: Methods and systems for defocusing a rendered computer-generated image are presented. Pixel values for a pixel array are determined from a scene description. A blur amount for each pixel is determined based on a lens function representing a lens shape and/or effect. A blur amount and blur transparency value are determined for the pixel based on the lens function and pixel depth. A convolution range comprising pixels adjacent to the pixel is determined based on the blur amount. A blend color value is determined for the pixel based on the color value of the pixel, color values of pixels in the convolution range, and the blur transparency value. The blend color value is scaled based on the blend color value and a modified pixel color value is determined from scaled blend color values.

Abstract: An apparatus to facilitate encoding of point cloud data is disclosed. The apparatus includes one or more processors to receive point cloud data including a plurality of images and camera parameters, generate encoded point cloud data including a color texture image and a depth image having cropped regions for each of the plurality of images, and metadata to describe the camera parameters and a mapping of the cropped regions to a real coordinate space.

Abstract: A data output apparatus includes: a video decoder that decodes a video stream to generate a first video signal; an external metadata acquisition unit that acquires one or more pieces of metadata corresponding to one or more first conversion modes; an HDR metadata interpreter that interprets one of the one or more pieces of metadata to acquire characteristic data and conversion auxiliary data; a DR converter that supports one or more second conversion modes and performs conversion processing of a luminance range of the first video signal based on the conversion auxiliary data to generate a second video signal; and an HDMI output unit that outputs the second video signal to a display apparatus.

Abstract: Disclosed are an image data processing method and a sensor device performing the same. The sensor device includes an image sensor configured to acquire image data, an image buffer configured to store the image data, and an image processor configured to generate image-processed data by applying a filter corresponding to a storage pattern of the image buffer to the image data stored in the image buffer.

Abstract: The color coding approach adopted by the Anchor Point Cloud Compression (PCC) leads to a significant color distortion, even at very low compression ratios. This color distortion is also referred to as color leaking. Two approaches are able to mitigate the color leaking. Both approaches reduce the color leaking significantly, while one of them (e.g., 1D subsampling) is more robust and preferred in most cases. A more general approach is an adaptive selection between these two approaches.

Abstract: An image signal processing device of the present disclosure includes a luminance correction section that performs, on a basis of information on a maximum output luminance value in a display section, luminance correction on an image signal to be supplied to the display section, the maximum output luminance value being variable.

Abstract: A method and a device for processing a current pixel block of an image using a palette prediction mode according to HEVC Format Range Extension. The mode uses a current palette to build a predictor block of indexes to predict the current pixel block. The current palette comprises entries associating entry indexes with pixel values. The method comprises predicting the current palette from entries of two or more palettes that are palettes previously used to process blocks of pixels. A palette predictor may be used that is built from the two or more palettes, preferably from the last used palette for a coding unit in the current coding entity and from the previously used palette for which a flag bitmap indicates whether or not its elements have been copied into the last used palette. Accordingly, the coding of the palette mode is improved.

Abstract: A computer is caused to realize: a line drawing data acquisition function to acquire line drawing data to be colored; a size-reducing process function to perform a size-reducing process on the line drawing data acquired to a predetermined reduced size so as to obtain size-reduced line drawing data; a first coloring process function to perform a coloring process on the size-reduced line drawing data based on a first learned model that has previously learned the coloring process on the size-reduced line drawing data by using sample data; and a second coloring process function to perform a coloring process on original line drawing data by receiving an input of the original line drawing data and colored, size-reduced line drawing data as the size-reduced line drawing data on which the first coloring process function has performed the coloring, based on a second learned model that has previously learned the coloring process on the sample data by receiving an input of the sample data and colored, size-reduced sampl

Abstract: An artificial intelligence (AI) decoding apparatus includes at least one processor configured to execute one or more instructions to: obtain a second image corresponding to a first image by performing first decoding on image data included in a main bitstream, obtain AI upscaling activation flag information included in AI data of a sub-bitstream, determine whether to perform AI upscaling on the second image, based on the AI upscaling activation flag information, when it is determined that AI upscaling is to be performed on the second image, obtain a third image by performing AI upscaling on the second image, through an upscaling deep neural network (DNN) set according to upscaling DNN information, the upscaling DNN information selected from among a plurality of pieces of pre-stored upscaling DNN information based on at least a portion of the image data and/or at least a portion of AI sub-data.

Abstract: A region extraction apparatus provided with a candidate extraction unit, a categorization unit and a region extraction unit. The candidate extraction unit extracts a difference region where a first ground surface image photographed from above at a first time point and a second ground surface image photographed at a second time point are different. The categorization unit estimates a first category of a first object photographed in correspondence with a position in the first ground surface image, calculates a first categorization image indicating a relationship between the position of the first object and the first category, estimates a second category of a second object photographed in correspondence with a position in the second ground surface image and calculates a second categorization image indicating a relationship between the position of the second object and the second category. The region extraction unit extracts an extraction region from the difference region.

Abstract: Disclosed is a process for producing an animation of the multipage type, in which each still image of the animation has scalability properties. Preferably, each still image is to be saved on at least two levels, the first level containing few details and each following level providing additional details to the preceding level.

Abstract: The disclosure regards cross-component prediction and methods for deriving of a linear model for obtaining a first-component sample for a first-component block from an associated reconstructed second-component sample of a second-component block in the same frame, the method comprising determining the parameters of a linear equation representing a straight line passing through two points, each point being defined by two variables, the first variable corresponding to a second-component sample value, the second variable corresponding to a first-component sample value, based on reconstructed samples of both the first-component and the second-component; and deriving the linear model defined by the straight line parameters; wherein said determining the parameters uses integer arithmetic.

Abstract: Automation controls and associated text in images displayed by a computer application are automatically detected by way of region-based R-FCN and Faster R-CNN engines. Datasets comprising images containing application controls, where the application controls include images of application where width is greater than height, width is equal to height and height is greater than width are retrieved and each dataset is processed with the R-FCN and Faster R-CNN engines to generate a software robot configured to recognize corresponding application controls. Text is recognized by an optical character recognition system that employs a deep learning system trained to process a plurality of images to identify images representing text within each image and to convert the images representing text to textually encoded data. The deep learning system is trained with training data generated from a corpus of real-life text segments that are generated by a plurality of OCR modules.

Abstract: A computer is caused to realize: a line drawing data acquisition function to acquire line drawing data to be colored; a size-reducing process function to perform a size-reducing process on the line drawing data acquired to a predetermined reduced size so as to obtain size-reduced line drawing data; a first coloring process function to perform a coloring process on the size-reduced line drawing data based on a first learned model that has previously learned the coloring process on the size-reduced line drawing data by using sample data; and a second coloring process function to perform a coloring process on original line drawing data by receiving an input of the original line drawing data and colored, size-reduced line drawing data as the size-reduced line drawing data on which the first coloring process function has performed the coloring, based on a second learned model that has previously learned the coloring process on the sample data by receiving an input of the sample data and colored, size-reduced sampl

Abstract: A method of video decoding performed in a video decoder. The method including receiving a coded video bitstream including a current picture. The method further including, performing an inverse quantization on a current block included in the current picture. The method further including performing, after performing the inverse quantization, an inverse transform on the current block. The method further including performing a prediction process on the current block after performing the inverse transform. The method further including, after performing the prediction process on the current block, determining whether a predetermined condition is satisfied. The method further including, in response to determining that the predetermined condition is met, performing an inverse color transform on the current block.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating video frames using neural networks. One of the methods includes processing a sequence of video frames using an encoder neural network to generate an encoded representation; and generating a predicted next frame pixel by pixel according to a pixel order and a channel order, comprising: for each color channel of each pixel, providing as input to a decoder neural network (i) the encoded representation, (ii) color values for any pixels before the pixel in the pixel order, and (iii) color values for the pixel for any color channels before the color channel in the channel order, wherein the decoder neural network is configured to generate an output defining a score distribution over a plurality of possible color values, and determining the color value for the color channel of the pixel by sampling from the score distribution.