Alibaba Group Holding Limited

Pixelworks, Inc.

Youku Information Technology (beijing) Co., Ltd.

Xiaojie Sheng

Neil Woodall

Guohua Cheng

Bob Zhang

Yue Ma

Yuxing Wu

Wei Li

Ji Wang

Wuping Du

Methods, apparatuses, mediums, and devices for generating multi-angle free-respective image data are provided. The method for generating multi-angle free-perspective image data includes: acquiring multiple synchronized images, where the multiple images have different shooting angles; determining the depth data of each image based on the multiple images; and for each of the images, storing pixel data of the image in a first field and storing depth data in a second field associated with the first field. Technical solutions in the example embodiments of the present invention may improve the user experience.

Method, apparatus, medium, and device for generating multi-angle free-respective image data

A method, a system, a device, and a computer readable storage medium for video reconstruction are disclosed. The method includes: obtaining an image combination of multi-angle free-perspective video frames, parameter data corresponding to the image combinations of the video frames, and position information of a virtual viewpoint based on a user interaction; selecting texture images and depth maps of corresponding groups in the image combinations of the video frames at a time moment of the user interaction according to a preset rule and based on the position information of the virtual viewpoint and the parameter data corresponding to the image combinations of the video frames; and combining and rendering the texture images and the depth maps of the corresponding groups based on the position information of the virtual viewpoint and parameter data corresponding to the depth maps and the texture images of the corresponding groups to obtain a reconstructed image.

Video reconstruction method, system, device, and computer readable storage medium

The disclosed embodiments relate to image processing methods and apparatuses. In one embodiment, a method includes: mapping an inputted three-dimensional (3D) model map into an asymmetric cubemap, the asymmetric cubemap located at a different place than the mapping center of the inputted 3D model map; and stretching the asymmetric cubemap mapped for the inputted 3D model map into a two-dimensional (2D) stretched plane map. Since the asymmetric cubemap is located at a different place than the mapping center of the inputted 3D model map, after the inputted 3D model map is mapped into the asymmetric cubemap and stretched into a 2D plane, the image resolution can be maintained in one face of the asymmetric cubemap and the image resolution can be reduced in other faces, so that high image resolution in an ROI of a user can be maintained, and the resolution of regions away from the ROI can slowly decrease, thereby lowering the overall resolution of the 2D stretched plane map and reducing the amount of data for subsequent processing.

Method and apparatus for image processing of three-dimensional model maps

Disclosed are video playback and data providing methods in a virtual scene, and a client and server implementing the same. The video playback method comprises: receiving current video segment data sent from a server, wherein the current video segment data represents a video segment, and the current video segment data comprises at least one specified viewing angle and a data identifier representing video segment data to which the specified viewing angle is directed; playing the video segment represented by the current video segment data, and acquiring a current viewing angle of a user during playback; and determining a target specified viewing angle matching the current viewing angle from the at least one specified viewing angle, wherein a video segment represented by video segment data to which the target specified viewing angle is directed is played at the end of the playback of the video segment represented by the current video segment data. The technical solution provided in the disclosure enables a user to move in a virtual scene while avoiding the use of external devices.

Providing video playback and data associated with a virtual scene

Image reconstruction methods, systems, devices, and computer-readable storage media are provided. The method includes: acquiring a multi-angle free-perspective image combination, parameter data of the image combination, and virtual viewpoint position information based on user interaction, where the image combination includes multiple groups of texture images and depth maps that are synchronized at multiple angles and have corresponding relationships; selecting a corresponding group of texture images and depth maps in the image combination at a user interaction moment based on a preset rule according to the virtual viewpoint position information and the parameter data of the image combination; and combining and rendering the selected corresponding group of texture images and depth maps in the image combination at the user interaction moment based on the virtual viewpoint position information and parameter data corresponding to the corresponding group of texture images and depth maps in the image combination at the user interaction moment.

Image reconstruction method, system, device and computer-readable storage medium

Panoramic video transcoding methods, devices, and apparatuses are provided. The method involves a central processing unit (CPU) and a graphics processing unit (GPU) and includes decoding a panoramic video; mapping, at least partly using the GPU, the decoded video data to N visual angles, to obtain N channels of video data, in which N is more than 1; encoding the N channels of video data respectively to obtain N channels of video streams; and slicing and packing the N channels of video streams respectively.

Methods, apparatuses and devices for panoramic video transcoding

Methods, apparatuses, media, and devices for generating multi-angle free-perspective video data are provided. The method for generating multi-angle free-perspective video data includes: acquiring multiple frame-synchronized videos, wherein shooting angles of the multiple frame-synchronized videos are different; parsing a respective video of the multiple frame-synchronized videos to obtain an image combination at a respective frame moment of multiple frame moments, wherein the image combination includes multiple frame-synchronized frame images; determining depth data of a respective frame image in the image combination based on the image combination at the respective frame moment; and generating a stitched image corresponding to the respective frame moment, wherein the stitched image includes a first field storing pixel data of the respective frame image in the image combination, and a second field storing the depth data of the respective frame image in the image combination.

Method, apparatus, medium, and server for generating multi-angle free-perspective video data

A method, an apparatus, a medium, a terminal, and a device for processing multi-angle free-perspective data are disclosed. The method includes: acquiring a data header file; determining a defined format of a data file according to a parsing result of the data header file; reading and obtaining a data combination from the data file based on the defined format, where the data combination includes pixel data and depth data of multiple synchronized images, and the multiple synchronized images have different perspectives with respect a to-be-viewed area, and pixel data and depth data of each image of the multiple synchronized images has an association relationship; and performing image or video reconstruction of a virtual viewpoint according to the read data combination, where the virtual viewpoint is selected from a multi-angle free-perspective range, and the multi-angle free-perspective range is a range supporting virtual viewpoint switching viewing of the to-be-viewed area.

Method, apparatus, medium, terminal, and device for processing multi-angle free-perspective data

A method, a computer storage medium, and a client for switching scenes of a panoramic video are provided. The method includes playing a first primary video stream in a first video stream group associated with a first perspective of a user; determining a video stream switching time when a perspective of the user is switched from the first perspective to a second perspective, the video stream switching time being associated with a time length of a video slice in a second backup video stream; and switching from the first primary video stream to the second backup video stream when a play time of the first primary video stream reaches the video stream switching time, so that content of the second backup video stream starting from the video stream switching time is played. The present application can improve the effects of switching between scenes of the panoramic video.

Method, computer storage media, and client for switching scenes of panoramic video

Embodiments of the present application provide a method and an apparatus, and a machine readable media for image processing. The method includes obtaining a panoramic video image, where the panoramic video image is determined based on a perspective mapping, and the panoramic video image includes a primary perspective region and at least one secondary perspective region; dividing the secondary perspective region into at least two sub-regions based on distribution information of high-frequency components in the secondary perspective region; determining respective filter templates of the sub-regions, and filtering the sub-regions using the respective filter templates; and determining a filtered panoramic video image.

Image processing method, apparatus and machine-readable media

Image mapping and processing methods, apparatuses and machine-readable media are disclosed. The methods include obtaining a panoramic video image, and demarcating the panoramic video image into a primary area and at least one ancillary area; determining a primary quantization parameter corresponding to the primary area, and an ancillary quantization parameter corresponding to the ancillary area, wherein the primary quantization parameter is smaller than the ancillary quantization parameter, with the quantization parameter being inversely proportional to an image quality; and encoding the primary area based on the primary quantization parameter, encoding the ancillary area based on the ancillary quantization parameter, and determining a coded panoramic video image.

Image mapping and processing method, apparatus and machine-readable media

A method of estimating motion in video display data includes dividing a frame of the video display data into blocks, generating at least two motion models for a current block from motion models of neighboring blocks to the current block, and selecting one of the motion models based upon a similarity of the motion model that minimizes differences between corresponding blocks in reference frames to the current block.

Motion estimation based on block level bilateral filter and local motion model

A method of generating motion vectors for image data includes identifying boundaries of at least one object in original frames of image data, performing object motion analysis based upon the boundaries, performing pixel-level motion layer generation, using the object motion analysis and the pixel-level motion layers to generate motion for blocks in the image data, and producing block level motion information and layer information for the original frames of image data.

True motion vector editing tool

To denoise video data, a temporal IIR filter is applied to video data. The temporally filtered frame is used as a guide to spatially filter the video data without introducing temporal artifacts. The spatially filtered frame can then be temporally filtered again using a temporal IIR filter. The results of the second temporal filtering can then be used as a guide to eliminate noise that the spatial filter did not eliminate.

Growing community of inventors

Shanghai, China

Xiaojie Sheng