Samsung Electronics Co., Ltd.

Hyundai Motor Company

Kia Corporation

Jong-seok Lee

Jae-hwan Kim

Young-o Park

Jeong-Hoon Park

Kwang-pyo Choi

Sun-young Jeon

Sang-Hyeok Yang

Ui-Yeon Won

Woong-il Choi

Dai-woong Choi

Dong-woo Kim

A-Rang Lee

A method for manufacturing a two-terminal memory device includes: forming an extended drain and a drain layer on a substrate; forming a ferroelectric layer covering the substrate and the extended drain; forming a semiconducting layer on the ferroelectric layer, and forming a source layer connected to the semiconducting layer on the ferroelectric layer.

Method for manufacturing a two-terminal memory device

A two-terminal memory device includes: a substrate; an extended drain extending from a drain and a lower surface of the drain and laminated on the substrate; a ferroelectric layer connected to the drain and covering the extended drain and the substrate; and a source laminated on the ferroelectric layer to face the drain.

Two-terminal memory device

A two-terminal memory device including: a substrate; a source and a drain formed to face each other on an upper surface of the substrate; a ferroelectric layer connected to the source and the drain and formed between the source and the drain; and an extended drain extending from the drain and laminated on the ferroelectric layer. The two-terminal memory device may be applied as a cross-point type and neuromorphic device capable of implementing multi-resistance levels with multi-layer switchable resistance layers.

Two-terminal memory device, a method for manufacturing the same, and a semiconductor device including a two-terminal memory device

Provided is in-loop filtering technology using a trained deep neural network (DNN) filter model. An image decoding method according to an embodiment includes receiving a bitstream of an encoded image, generating reconstructed data by reconstructing the encoded image, obtaining information about a content type of the encoded image from the bitstream, determining a deep neural network (DNN) filter model trained to perform in-loop filtering by using at least one computer, based on the information about the content type, and performing the in-loop filtering by applying the reconstructed data to the determined DNN filter model.

Method and device for encoding or decoding image

Provided is an image encoding or decoding method and device capable of increasing image compression efficiency by determining a differential quantization parameter of a current block based on statistical information of original samples of the current block or samples reconstructed previously to the current block. The image decoding method includes receiving a bitstream, obtaining a predicted quantization parameter value and a quantized transformation coefficient of a current block from the bitstream, obtaining a differential quantization parameter of the current block based on statistical information of samples reconstructed previously to the current block, obtaining a quantization parameter of the current block based on the differential quantization parameter of the current block and the predicted quantization parameter value, and inversely quantizing the quantized transformation coefficient of the current block based on the quantization parameter of the current block.

Video encoding method and device and video decoding method and device

Provided is a prediction image generating technology using a deep neural network (DNN). Provided is an image decoding method including: receiving a bitstream of an encoded image; determining at least one block split from the encoded image; determining neighboring blocks for predicting a current block among the at least one block; generating prediction data of the current block by applying the neighboring blocks to a DNN learning model configured to predict a block of an image by using at least one computer; extracting residual data of the current block from the bitstream; and reconstructing the current block by using the prediction data and the residual data.

Provided are an image compressing method including determining a compressed image by performing downsampling using a deep neural network (DNN) on an image; determining a prediction signal by performing prediction based on the compressed image; determining a residual signal based on the compressed image and the prediction signal; and generating a bitstream comprising information about the residual signal, wherein the DNN has a network structure that is predetermined according to training of a downsampling process using information generated in an upsampling process, and an image compressing device for performing the image compressing method. Also, provided are an image reconstructing method of reconstructing a compressed image by using a DNN for upsampling, the compressed image having been compressed by the image compressing method, and an image reconstructing device for performing the image reconstructing method.

Method for encoding/decoding image and device therefor

Provided is a method of encoding an image, the method including: determining a subjective quality of the image when the image is compressed; determining at least one degree of compression that changes the subjective quality and is from among degrees of compression indicating how much the image is compressed; and encoding the image by compressing a residual signal of the image, based on compression information according to the determined degree of compression, wherein the subjective quality is determined for each frame by using a Deep Neural Network (DNN). Provided are an image decoding method and an image decoding apparatus for performing the image decoding method for decoding an image by using information encoded according to an image encoding method.

Method for encoding/decoding image, and device therefor

A method of determining whether an image is to be re-encoded includes obtaining a first quantization matrix from an image file including the image, the image being encoded by quantization based on the first quantization matrix including a plurality of first quantization parameters; obtaining a second quantization matrix from a re-encoding device, the second quantization matrix including a plurality of second quantization parameters and having the same size as the first quantization matrix; determining a comparison coefficient based on elements greater than '0' among elements of a comparison matrix obtained by subtracting the first quantization matrix from the second quantization matrix; and determining that the image is to be decoded by inverse quantization based on the first quantization matrix and the decoded image is to be re-encoded by quantization based on the second quantization matrix, when the comparison coefficient is greater than a first threshold value.

Growing community of inventors

Suwon-si, South Korea

Jong-seok Lee