Abstract: The present invention relates to a method and apparatus for encoding a displacement video using image tiling. A method for encoding multi-dimensional data according to an embodiment of the present disclosure may comprise: converting the multi-dimensional data into one or more frames with two-dimensional characteristics; generating one or more frame groups by grouping the one or more frames with pre-configured number units; reconstructing frames belonging to each frame group into a tiled frame; and generating a bitstream by encoding the tiled frame. Here, the tiled frame may be constructed with one or more blocks, and each block may be constructed by rearranging pixels existing at the same location in the frames.

Abstract: Disclosed herein are a method, an apparatus and a storage medium for image encoding/decoding using a segmentation map. A feature vector for an image may be extracted using a segmentation map. The image may be encoded using the segmentation map and the feature vector. An output stream from an encoding apparatus may include a video stream and a feature stream. An input stream to a decoding apparatus may include a video stream and a feature stream. The image may be reconstructed using a reconstructed segmentation map and a reconstructed feature vector.

Abstract: Disclosed herein are an apparatus and method for detecting a keypoint based on deep learning robust to scale changes based on information change across receptive fields. The apparatus for detecting a keypoint based on deep learning robust to scale changes based on information change across receptive fields includes a feature extractor for extracting a feature from an input image based on a pre-trained deep learning neural network, an information accumulation pyramid module for outputting, from the feature, at least two filter responses corresponding to receptive fields having different scales, an information change detection module for calculating an information change between the at least two filter responses, a keypoint detection module for creating a score map having a keypoint probability of each pixel based on the information change, and a continuous scale estimation module for estimating a scale of a receptive field having a biggest information change for each pixel.

Abstract: The present invention relates to a method and apparatus for encoding a displacement video using image tiling. A method for encoding multi-dimensional data according to an embodiment of the present disclosure may comprise: converting the multi-dimensional data into one or more frames with two-dimensional characteristics; generating one or more frame groups by grouping the one or more frames with pre-configured number units; reconstructing frames belonging to each frame group into a tiled frame; and generating a bitstream by encoding the tiled frame. Here, the tiled frame may be constructed with one or more blocks, and each block may be constructed by rearranging pixels existing at the same location in the frames.

Abstract: An encoding/decoding apparatus and method for controlling a channel signal is disclosed, wherein the encoding apparatus may include an encoder to encode an object signal, a channel signal, and rendering information for the channel signal, and a bit stream generator to generate, as a bit stream, the encoded object signal, the encoded channel signal, and the encoded rendering information for the channel signal.

Abstract: The present invention discloses an encoding apparatus using a Discrete Cosine Transform (DCT) scanning, which includes a mode selection means for selecting an optimal mode for intra prediction; an intra prediction means for performing intra prediction onto video inputted based on the mode selected in the mode selection means; a DCT and quantization means for performing DCT and quantization onto residual coefficients of a block outputted from the intra prediction means; and an entropy encoding means for performing entropy encoding onto DCT coefficients acquired from the DCT and quantization by using a scanning mode decided based on pixel similarity of the residual coefficients.

Abstract: Disclosed is an apparatus and method for processing a multichannel audio signal. A multichannel audio signal processing method may include: generating an N-channel audio signal of N channels by down-mixing an M-channel audio signal of M channels; and generating a stereo audio signal by performing binaural rendering of the N-channel audio signal.

Abstract: Disclosed herein are an apparatus and method for searching for a global minimum of a point cloud registration error. The apparatus includes memory in which at least one program is recorded and a processor for executing the program. The program performs collecting multiple registration results in which the registration error between a source point cloud and a target point cloud is a local minimum as candidates and selecting the registration result in which the registration error between the source point cloud and the target point cloud is a global minimum, among the candidates. Collecting the multiple registration results may comprise repeatedly randomly initializing the source point cloud and the target point cloud and registering the initialized source point cloud to the initialized target point cloud to thereby search for a registration result in which the registration error therebetween is a local minimum.

Abstract: An encoding/decoding apparatus and method for controlling a channel signal is disclosed, wherein the encoding apparatus may include an encoder to encode an object signal, a channel signal, and rendering information for the channel signal, and a bit stream generator to generate, as a bit stream, the encoded object signal, the encoded channel signal, and the encoded rendering information for the channel signal.

Abstract: Disclosed is an apparatus and method for processing a multichannel audio signal. A multichannel audio signal processing method may include: generating an N-channel audio signal of N channels by down-mixing an M-channel audio signal of M channels; and generating a stereo audio signal by performing binaural rendering of the N-channel audio signal.

Abstract: Disclosed herein are an apparatus and method for sensing an image based on an event. The apparatus includes memory in which at least one program is recorded and a processor for executing the program. The program may perform acquiring at least one of brightness information and color information from an input image signal, performing conversion including at least one of filtering of at least one of the acquired brightness information and color information, color conversion, and brightness conversion, calculating a quantized difference between a first converted image converted from the currently input image signal and a second converted image converted from a previously input image signal, and generating event information for the input image signal as a bitstream based on the quantized difference.

Abstract: Provided is an apparatus and method for encoding/decoding a moving picture based on adaptive scanning. The moving picture apparatus and method can increase a compression rate based on adaptive scanning by performing intra prediction onto blocks of a predetermined size, and scanning coefficients acquired from Discrete Cosine Transform (DCT) of a residue signal and quantization differently according to the intra prediction mode. The moving picture encoding apparatus includes: a mode selector for selecting and outputting a prediction mode; a predictor for predicting pixel values of pixels to be encoded of an input video based on the prediction mode to thereby output a residue signal block; a transform/quantization unit for performing DCT onto the residue signal block and quantizing the transformed residue signal block; and an encoder for adaptively scanning and encoding the quantized residue signal block based on the prediction mode.

Abstract: An immersive image encoding method according to the present disclosure includes classifying a plurality of view images into a basic image and an additional image; performing pruning for at least one of the plurality of view images based on the classification result; generating a depth atlas based on a result of performing the pruning; and correcting an occupancy state of pixels in the depth atlas.

Abstract: Provided is an apparatus and method for encoding/decoding a moving picture based on adaptive scanning. The moving picture apparatus and method can increase a compression rate based on adaptive scanning by performing intra prediction onto blocks of a predetermined size, and scanning coefficients acquired from Discrete Cosine Transform (DCT) of a residue signal and quantization differently according to the intra prediction mode. The moving picture encoding apparatus includes: a mode selector for selecting and outputting a prediction mode; a predictor for predicting pixel values of pixels to be encoded of an input video based on the prediction mode to thereby output a residue signal block; a transform/quantization unit for performing DCT onto the residue signal block and quantizing the transformed residue signal block; and an encoder for adaptively scanning and encoding the quantized residue signal block based on the prediction mode.

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 of processing an immersive video according to the present disclosure includes performing pruning for an input image, generating an atlas based on patches generated by the pruning and generating a cropped atlas by removing a background region of the atlas.

Abstract: Disclosed herein is a method of creating an image stitching workflow including acquiring 360-degree virtual reality (VR) image parameters necessary to makes a request for image stitching and create the image stitching workflow, acquiring a list of functions applicable to the image stitching workflow, creating the image stitching workflow based on functions selected from the list of functions, determining the number of media processing entities necessary to perform tasks configuring the image stitching workflow and generating a plurality of media processing entities according to the determined number of media processing entities, and allocating the tasks configuring the image stitching workflow to the plurality of media processing entities.

Abstract: Disclosed is a binaural rendering method and apparatus for decoding a multichannel audio signal. The binaural rendering method may include: extracting an early reflection component and a late reverberation component from a binaural filter; generating a stereo audio signal by performing binaural rendering of a multichannel audio signal base on the early reflection component; and applying the late reverberation component to the generated stereo audio signal.

Abstract: A power management integrated circuit includes a reconfigurable converter connected to a rechargeable battery and a control circuit. The reconfigurable converter includes an indcutor to accumulate energy based on an input voltage from a thermoelectric device, provides a wireless sensor node connected to an output node with an output voltage based on the input voltage and an internal input voltage of the rechargeable battery, operates in one of operation modes based on an available power provided from the thermoelectric device and a power consumption by the wireless sensor node and drives the wireless sensor node in response to the wireless sensor node consuming a current greater than a maximum current generated by the thermoelectric device. The control circuit to control the reconfigurable converter based on the input voltage, a first voltage of a first switching node, a second voltage of a second switching node and the output voltage.

Abstract: Disclosed herein are an apparatus and method for detecting a keypoint based on deep learning robust to scale changes based on information change across receptive fields. The apparatus for detecting a keypoint based on deep learning robust to scale changes based on information change across receptive fields includes a feature extractor for extracting a feature from an input image based on a pre-trained deep learning neural network, an information accumulation pyramid module for outputting, from the feature, at least two filter responses corresponding to receptive fields having different scales, an information change detection module for calculating an information change between the at least two filter responses, a keypoint detection module for creating a score map having a keypoint probability of each pixel based on the information change, and a continuous scale estimation module for estimating a scale of a receptive field having a biggest information change for each pixel.