Sony Corporation

Sony Pictures Entertainment, Inc.

Hitachi-kokusai Electric Inc.

Sony Group Corporation

Sony Corporation of America

Panasonic Corporation

Sony Pictures Technologies Inc.

Nisseitechnica Co., Ltd.

Yoshikazu Takashima

Kenjiro Ueda

Katsumi Muramatsu

Tateo Oishi

Motoki Kato

Miaoqi Zhu

Jun Yonemitsu

Tomoshi Taniyama

Tomoyuki Asano

Yoshitomo Osawa

Shinobu Hattori

Shinya Morita

Takashi Ozaki

Kiyohiko Maeda

Satoshi Kitani

Saravana Prabhu Anbananthan

Sadao Hisakado

Hiroshi Unami

Yoshiyuki Kobayashi

Kouichi Uchimura

Masanobu Nakamura

Koji Obata

Nobuhiro Igi

Mitsutoshi Terada

Michael K Thompson

Christopher Taylor

Michael Lafuente

Mengyu Chen

Stephen Shapiro

Ouyang Chao

Daniel De La Rosa

Masayuki Kozuka

Masaya Yamamoto

Toshihisa Nakano

Kaoru Murase

Toshiya Hamada

Kazuo Yamamoto

Kenichi Suzaki

Ayako Watanabe

Minehisa Nagata

Seiyo Nakashima

Senichi Onoda

Tommy Choy

Dennis Adams

Takashi Yamanishi

Spencer Stephens

Don Charles Eklund, Ii

John B Ying

Robert Aubey, Jr

Peter Avventi

George Reis

Kiruthika Krishnadevarajan

Kenjiro Uedo

Mikio Ohno

Improving the accuracy of predicted segmentation masks, including: extracting a ground-truth RGB image buffer and a binary contour image buffer from a ground-truth RGB image container for segmentation training; generating predicted segmentation masks from the ground-truth RGB image buffer; generating second binary contours from the predicted segmentation masks using a particular algorithm; computing a segmentation loss between manually-segmented masks of the ground-truth RGB image buffer and the predicted segmentation masks; computing a contour accuracy loss between contours of the binary contour image buffer and the binary contours of the predicted segmentation masks; computing a total loss as a weighted average of the segmentation loss and the contour accuracy loss; and generating improved binary contours by compensating the contours of the binary contour image buffer with the computed total loss, wherein the improved binary contours are used to improve the accuracy of the predicted segmentation masks.

Machine learning based image segmentation training with contour accuracy evaluation

Enhancing color reproduction of an image in an upscaling process, the method comprising: converting RGB-formatted data into luminance channel formatted data and color channel formatted data; converting RGB-predicted data into luminance channel predicted data and a color channel predicted data; computing a first loss between the RGB-formatted data and the RGB-predicted data; computing a second loss between the color channel formatted data and the color channel predicted data; and outputting a weighted average value of the first loss and the second loss to enhance the color reproduction of the image.

Enhanced color reproduction for upscaling

Methods and systems including: applying a gamma adjustment to frames of an input video with particular values of parameters to generate at least one set of gamma-adjusted frames to generate at least one set of gamma-adjusted frames; applying a segmentation technique to the at least one set of gamma-adjusted frames to generate segmentation masks; applying an optical flow technique to the at least one set of the gamma-adjusted frames to generate optical flow maps; and combining the segmentation masks and the optical flow maps to generate hybrid segmentation masks.

Hybrid video segmentation aided by optical flow for objects in motion

Frame handling an input image in an ML-based upscaling process to produce an output image, including: detecting and extracting boxing edge regions from edges of an active picture area of the input image; extending the extracted boxing edge regions by adding extra pixels around each of the boxing edge regions to produce padded edge regions; upscaling the padded edge regions to produce upscaled padded edge regions; cropping each of the upscaled padded edge regions to a size corresponding to the size of each of the boxing edge regions prior to the upscaling multiplied by an upscaling factor; inserting the cropped edge regions into the edges of the output image; and upscaling a non-edge region of the input image and inserting the upscaled non-edge region into the output image.

Frame handling for ML-based upscaling

Enabling adjustment of sharpness and details of an input image in upscaling, including: applying a Fourier transform function on a brightness channel of the input image to generate a 2-D frequency map; adjusting the 2-D frequency map to control a target amount of sharpness and details in an upscaled output image; and using the adjusted 2-D frequency map as an additional input channel along with standard color image data for a training and upscaling process.

Adjusting sharpness and details in upscaling output

Training and upscaling a large-sized input image, including: dividing the large-sized input image into a plurality of small-sized sub-pictures; expanding each sub-picture of the plurality of small-sized sub-pictures using target padding pixels to produce an expanded sub-picture; upscaling each sub-picture using an ML-based upscaler to produce an expanded upscaled sub-picture; cropping the expanded upscaled sub-picture to an upscaled size equal to an original size of each sub-picture multiplied by an upscaling factor; repeating expanding, upscaling, and cropping for the plurality of sub-pictures; and concatenating the plurality of cropped sub-pictures to produce an output image.

Training and upscaling of large size image

Content conversion verification, includes: converting, at a first computer system, an original content file to a target format, generating a converted content file in the target format; generating a checksum for the converted content file; and sending the original content file and the checksum to a second computer system. Key words include content verification and checksum.

Content format conversion verification

A method for enhancing color reproduction in an upscaling process, including: converting RGB-formatted data to color-space-separable-formatted data; sending the RGB-formatted data to a neural network for training to generate RGB-predicted data; converting the RGB-predicted data to color-space-separable-predicted data; computing a first loss function by calculating a first difference between the RGB-formatted data and the RGB-predicted data; extracting color-formatted data from the color-space-separable-formatted data; extracting color-predicted data from the color-space-separable-predicted data; computing a second loss function by calculating a second difference between the color-formatted data and the color-predicted data; and adjusting a balance between the first and second loss functions.

Content individualization, including: encrypting a first part of a source data set using a first key creating a first encrypted data set; encrypting a second part of the source data set using a second key creating a second encrypted data set; encrypting the second part of the source data set using a third key creating a third encrypted data set; and combining the first encrypted data set, the second encrypted data set, and the third encrypted data set to form a final encrypted data set. Key words include watermarking and content individualization.

Content individualization

An information processing device includes: a data processing unit which executes a reproduction process of recorded data of a disc, in which the data processing unit reads disc type information that is recorded on the disc, and ascertains a recorded data configuration of the disc based on the disc type information that is read, and executes data reading-out and reproduction processes according to a reading-out rate corresponding to the ascertained recorded data configuration.

Information processing device, information recording medium, information processing method, and program

Provided is an information processing device including: a data processing unit that performs a content playback process; and a communication unit that communicates with a server, wherein the content is content that has a segment region configured with a plurality of pieces of variation data which can be respectively decrypted by different keys, and for which a plurality of playback paths can be configured according to variation data which is selected as a playback object, wherein each piece of variation data is data for which a variation data identifier can be analyzed from decrypted data, and wherein the data processing unit receives playback path information which is allowed to be played from the server through the communication unit, and selects variation data from the segment region so as to perform content playback, according to the playback path information received from the server.

Information processing device, information processing method, and program

Provided is an information processing device including: a data processing unit that performs format conversion of converting MPEG-2 TS format data into MP4 format data, wherein the MPEG-2 TS format data is data that includes each segment region having a plurality of pieces of variation data decryptable with different keys, and wherein the data processing unit selects encrypted MP4 variation data from the segment region of the MPEG-2 TS format data and stores the MP4 variation data as MP4 format structure data in an MP4 file, while the encrypted variation data remains, without performing a decryption process and a re-encryption process for the selected MP4 variation data.

Using multimedia data for both continuous playback and streaming playback purposes, including: encoding an elementary stream of multimedia content into a fixed length of a group of pictures; multiplexing the encoded multimedia content into a transport stream; locating data offsets of each GOP header inside the transport stream; and uploading the data offsets along with the transport stream to a server. Continuous and streaming playback of multimedia content.

Dual-type of playback for multimedia content

A configuration for achieving efficient content verification processing based on hash values is provided. Hash values of hash units set as segmented data of a content stored on an information storage medium are recorded in a content hash table and are stored on the information storage medium together with the content. An information processing apparatus for executing content playback executes hash-value comparison processing based on one or more randomly selected hash values. Regardless of the data amount of content, the configuration can perform hash-value determination and comparison processing based on hash units having a small amount of data, so that user equipment for executing content playback can perform efficient content verification.

Information processing apparatus, information storage medium, content management system, data processing method, and computer program

Growing community of inventors

Tokyo, Japan

Yoshikazu Takashima