Abstract: A method, decoder, and apparatus are provided. Responsive to a current block being a MIP predicted block, it is determined whether it has one or multiple transform blocks. A MIP weight matrix to be used to decode the current block is determined based on a MIP prediction mode. Responsive to the MIP predicted block having one transform block, the MIP predicted block is derived based on the MIP weight matrix and previously decoded elements in the bitstream. Responsive to the MIP predicted block having multiple transform blocks: deriving a first MIP predicted block is derived based on the MIP weight matrix and previously decoded elements in the bitstream and remaining MIP predicted blocks are derived based further on decoded elements in at least one decoded transform block of the current block. The MIP predicted block(s) are output for subsequent processing.

Abstract: Methods and systems for detecting motion using video data from a smart camera device. In some embodiments, and by non-limiting example, this disclosure is directed to detecting motion in the compressed video domain. One aspect is a method comprising receiving compressed video data, extracting macroblocks and motion vectors for a plurality of frames in the compressed video data, identifying frame-level features for each of the plurality of frames based on the macroblocks and the motion vectors, calculating similarity features for each of the identified frame-level features based on the frame-level features identified in consecutive frames, and predicting motion for each of the plurality of frames by providing the frame-level features and the similarity features into a model trained to detect motion.

Abstract: The technology described in this document can be embodied in systems and computer-implemented methods for determining a score representing an amount of staining of the cornea. The methods include obtaining a digital image of the cornea stained with a tracer material, receiving a selection of a portion of the image, and processing, by a processing device, the selection to exclude areas with one or more artifacts to define an evaluation area. For each of a plurality of pixels within the evaluation area, a plurality of Cartesian color components are determined and a hue value in a polar coordinate based color space is calculated from the components. An amount of staining of the cornea is then determined as a function of the hue value. The methods also include assigning a score to the evaluation area based on the amount of staining calculated for the plurality of pixels.

Abstract: The embodiments disclosed herein relate to the construction of a search space for determining a motion vector for a current block of a picture in a video sequence. The search space construction is split into in two main stages, wherein a first and a second partial search space are respectively determined. Based on an initial estimate of a motion vector, a first search space is first constructed. A first and a second candidate motion of the first search space are identified according to a cost function. Based on the first and the second candidate motion vectors, a second search space is constructed. The motion vector for the current block is selected from the candidate motion vectors of the first search space and the second search space.

Abstract: A method and an active commissioning system for video analytics (VA) of video surveillance are provided. The method includes capturing, using a video capture device, a video sequence and receiving, at a VA processor, a parameter set from a cloud resource, wherein the parameter set is used for video analytics. The method also includes processing the video sequence by selecting subparts of the video sequence as events, wherein the events are detected based on the parameter set, generating a visualized report using the events from the VA processor and providing the visualized report to an operator, marking, by the operator, yes/no on the events in the visualized report, generating annotations for events based on the marking, and transmitting the annotations and the events to the cloud resource for processing, wherein the cloud resource processes the annotations and events and provides the parameter set based on the processing.

Abstract: To improve a coding efficiency. There are included a PU level search unit configured to search for a motion vector for each prediction block by using a matching process. and a sub-block level search unit configured to search for a motion vector of each of sub-blocks in the prediction block, wherein a precision of a local search by the PU level search unit is lower than a precision of a local search by the sub-block level search unit.

Abstract: A method for decoding a 360-degree image includes: receiving a bitstream obtained by encoding a 360-degree image; generating a prediction image by making reference to syntax information obtained from the received bitstream; combining the generated prediction image with a residual image obtained by dequantizing and inverse-transforming the bitstream, so as to obtain a decoded image; and reconstructing the decoded image into a 360-degree image according to a projection format. Here, generating the prediction image includes: checking, from the syntax information, prediction mode accuracy for a current block to be decoded; determining whether the checked prediction mode accuracy corresponds to most probable mode (MPM) information obtained from the syntax information; and when the checked prediction mode accuracy does not correspond to the MPM information, reconfiguring the MPM information according to the prediction mode accuracy for the current block.

Abstract: A load in processing of searching for a motion vector is reduced. In order to solve the problem described above, a motion vector derivation apparatus according to one aspect of the present invention that derives a motion vector to be referred to for generating a prediction image to be used for coding or decoding of a video includes a motion vector search unit configured to search for a motion vector on a prediction unit basis through matching processing. The motion vector search unit is configured to stop search of the motion vector, depending on whether or not a conditional expression according to a pixel bit-depth is satisfied.

Abstract: A method for video processing is provided. The method includes determining, based on a coding mode of a parent coding unit of a current coding unit that uses an affine coding mode or a rate-distortion (RD) cost of the affine coding mode, a usage of an adaptive motion vector resolution (AMVR) for a conversion between a coded representation of a current block of a video and the current block; and performing the conversion according to a result of the determining.

Abstract: Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for data indexing. The method for data indexing includes obtaining a plurality of data blocks for storage. The method further includes generating a plurality of vectorized representations of the plurality of data blocks so that a difference between vectorized representations of data blocks with a higher similarity among the plurality of data blocks is smaller. The method further includes storing the plurality of vectorized representations for indexing the plurality of data blocks respectively. The vectorized representations serve as indexes of the data blocks, and the vectorized representations may be used to effectively measure similarities between the data blocks, thereby providing the possibility of performing a fuzzy search of the data blocks based on the similarities.

Abstract: Disclosed is an inter prediction method which includes deriving control points (CPs) for the current block, wherein the CPs include a first CP and a second CP, deriving a first motion vector predictor (MVP) for the first CP and a second MVP for the second CP based on neighboring blocks of the current block, decoding a first motion vector difference (MVD) for the first CP, decoding a difference of two MVDs (DMVD) for the second CP, deriving a first motion vector (MV) for the first CP based on the first MVP and the first MVD, deriving a second MV for the second CP based on the second MVP and the DMVD for the second CP, and generating a predicted block for the current block based on the first MV and the second MV.

Abstract: The precision of up-sampling operations in a layered coding system is preserved when operating on video data with high bit-depth. In response to bit-depth requirements of the video coding or decoding system, scaling and rounding parameters are determined for a separable up-scaling filter. Input data are first filtered across a first spatial direction using a first rounding parameter to generate first up-sampled data. First intermediate data are generated by scaling the first up-sampled data using a first shift parameter. The intermediate data are then filtered across a second spatial direction using a second rounding parameter to generate second up-sampled data. Second intermediate data are generated by scaling the second up-sampled data using a second shift parameter. Final up-sampled data may be generated by clipping the second intermediate data.

Abstract: A loss function of a signal including an audio signal is determined. A loss function determining system for an audio signal is provided. A loss function is determined by: determining a reference quantization index by quantizing an original input signal; inputting the original input signal to a neural network classifier and applying an activation function to an output layer of the neural network classifier; and determining a total loss function for the neural network classifier using an output of the activation function and the reference quantization index.

Abstract: An apparatus and method determining a motion vector for a block, which relates to motion vector determination employing template matching or bilateral matching, are provided. At first, candidate motion vectors are obtained and are mapped to respective representatives (representative motion vectors). In particular, if the obtained candidate motion vectors point to fractional positions and thus, interpolation from the integer sample positions is required, the initial motion vector is rounded to a closest integer sample position. The rounded position is then used to define the search space for the template or bilateral matching to determine the best matching motion vector for the block (in other words, a refinement of an initial motion vector around which the candidate space is spanned). Finally, the mapping operation can be reverted in order to obtain a result in the original search space.

Abstract: A measurement apparatus comprising an acquisition memory adapted to store data sections of at least one acquired measurement signal; a processor adapted to calculate a measurement parameter vector, v, for each data section of the acquired measurement signal; and a trained autoencoder neural network adapted to process the measurement parameter vectors, v, applied as input data to the trained autoencoder neural network to provide at a middle layer of said autoencoder neural network an encoded vector, h, with characteristic signal features of the acquired measurement signal.

Abstract: A new real-time video communication system and method improves the clarity of shared screen by automatically adjusting the key frame's quantization parameter values. The adjustment depends on the underlying network's connection condition. When the connection bandwidth deteriorates, the key frame's quantization parameter value is further decreased on the value output from a standard H.264 encoder to derive a desirable quantization parameter value and thus achieve a desirable user experience of the shared screen. Moreover, the new real-time video communication system and method adjusts the frame rate of the shared screen based on the present quantization parameter value to achieve a more desirable user experience of the screen sharing.

Abstract: A RAM_HIME used for integer pixel motion estimation by an IME stores integer pixel luminance data from a SDRAM while satisfying the conditions that improve efficiency in reading an extracted rectangular area. A RAM_HSME used for motion estimation of quarter-pixel accuracy by a SME stores partial quarter-pixel luminance data while satisfying the conditions that improve efficiency in obtaining a rectangular area after calculation by calculation. A RAM_HMEC used for chrominance data generation of quarter-pixel accuracy by a QPG stores integer pixel chrominance data from the SDRAM while satisfying the conditions that improve efficiency in obtaining rectangular areas after calculation by calculation.

Abstract: Local motion estimation is described herein. Each picture of a video is partitioned into blocks for the local motion estimation. An extended-block FFT is calculated for each block, where the extended-block denotes that a certain area around the block is also included for applying FFT. Extending the block for FFT helps to account for the motion of objects that are moving into or out of the block. Phase correlation is applied to attain a set of Motion Vector (MV) candidates for the blocks, and a cost function is evaluated for each MV. If no MV candidate produces a cost function below a pre-defined threshold, a hierarchical variable block matching search is applied and the process is repeated with blocks for finer resolution. Also, predictive MV candidates are used during the block matching search along with temporal constraints tracking to select an MV that yields the minimum cost function.