Thomson Licensing

Realsee (beijing) Technology Co., Ltd.

Other

Ke.com (beijing)technology Co., Ltd.

Interdigital Ce Patent Holdings

Xiaodong Gu

Fan Zhang

Zhibo Chen

Ning Liao

Debing Liu

Feng Xu

Cihui Pan

Mingyuan Wang

Chenglin Liu

Wenbo Shi

Yonglin Yang

Kai Xie

Tong Rao

Chuanming Wang

Guoping Qiu

Quqing Chen

A method includes obtaining a color image and a first depth image; determining a target area of the first depth image, the target area of the first depth image having a first set of pixels with a first set of depth values; based on comparing an accuracy of the first set of depth values with a preset accuracy threshold: determining a target area of the color image that corresponds to the target area of the first depth image; determining, based on the target area of the color image, a second set of pixels with a second set of depth values for the target area of the first depth image; generating a second depth image that replaces the first set of pixels of the first depth image with the second set of pixels with the second set of depth values; outputting the second depth image as a three dimensional image.

Method and device for depth map completion

A method for generating an image of a scene with a corresponding depth map is disclosed herein. The method comprises collecting a plurality of copies of a two-dimensional image of the scene, generating an intermediate image by concatenating the plurality of copies of the two-dimensional image along a first direction, generating an intermediate depth map corresponding to the intermediate image by applying a neural network to the intermediate image, and generating, from the intermediate image and the intermediate depth map, the image of the scene with the corresponding depth map.

Method and device for generating virtual reality data

A method for combining multiple images is disclosed herein. A target mapping matrix is determined based on a first image and a second image. The target mapping matrix is associated with a target correspondence between the first image and the second image. The first image and the second image are combined into a combined image based on the first target mapping matrix. The combined image is output by implementing the disclosed method.

Method for merging multiple images and post-processing of panorama

A particular implementation determines objective distortion levels (d,) respectively for a plurality of artifact types. The objective distortion levels are aligned to ensure that the same distortion level of different types of artifacts corresponds to the same perceived distortion level. The aligned distortion levels (d,') are sorted to obtain sorted distortion levels (d,&#x2033;). The sorted distortion levels are then pooled together into an overall distortion level or an overall quality metric. The sorted distortion levels may be pooled using a weighted sum, wherein the weight is larger when the sorted distortion level is greater.

Video quality measurement considering multiple artifacts

The present invention generally relates to automatic detection of video artifacts. The present invention relates to automatically detect out the images with visible artifacts due to STB decoding error based on a generated artificial edge map. In particular, the application teaches a method to detect out the images with the visible artifacts and how to identify artificial edges from texture (natural) edges. The present system identifies that artificial edges are most significant in local area, while texture edges are often together with some other texture edges.

Method and apparatus to detect artificial edges in images

Various implementations address distortion and quality measurements related to, for example, freeze-with-skip and/or a freeze-without-skip events. In several implementations, information is accessed indicating that a first and second set of one or more consecutive pictures are not to be displayed. A first and second indicator are determined. In one such implementation, the first and second indicators indicate an amount of distortion across one or more types of distortive effects that result from displaying substantially a first, or second, same picture during a display time for the first, or second, set of pictures. The first and second indicators are combined in a non-linear manner. In another such implementation, the distortion is associated with a given type of distortive effect, from among multiple types of distortive effects, and the first and second indicators are combined for the given type of distortive effect.

Distortion/quality measurement

An image is partitioned into a foreground area, a background area, and optionally a transitional area. The partitioning may be pre-defined, or it may be based on user inputs and configuration data. The partitioning may also be refined based on an initial partitioning. Blur measures are determined respectively for the partitioned areas. A blur measure for the whole image can then be determined from a weighted average of the blur measures for the partitioned areas. The blur measure for the image can be used in a video quality monitor.

Blur measurement

A method for processing an image divided into blocks of pixels is described. The method includes detecting a largest sub-block whose pixels have an equal luminance value and identifying if the detected largest sub-block is or is not a natural texture. Thereafter, a weighted luminance difference is calculated between the detected largest sub-block and neighboring pixels and a block blockiness level is determined on the basis of the number of pixels within the detected largest sub-block, of the identification step and of the weighted luminance difference. The image is then processed on the basis of the block blockiness level.

Method of processing an image based on the determination of blockiness level

Various implementations address distortion and quality measurements. Certain implementations relate to distortions resulting in a freeze-with-skip and/or a freeze-without-skip. In one implementation, a pause term provides an indication of distortion associated with a pause arising from not displaying one or more consecutive pictures that are not to be displayed. A motion term provides an indication of distortion associated with a related change in motion. A combined indicator is based on an addition of the pause term and the motion term. In another particular implementation, a first indicator indicates distortion across one or more types of distortive effects that results from not displaying a first set of pictures. A second indicator indicates corresponding distortion with respect to a second set of pictures. The first indicator and the second indicator are combined in a non-linear manner.

The invention is made in the technical field of video quality measurement. More precisely, the invention is related to mean observer score prediction using a trained semi-supervised learning regressor. That is, a method and apparatus for measuring video quality using a semi-supervised learning system for mean observer score prediction is proposed. Said semi-supervised learning system comprises at least one semi-supervised learning regressor and said method comprises training the learning system and retraining the trained learning system using a selection of test data wherein the test data is used for determining at least one mean observer score prediction using the trained learning system and the selection is indicated by a feedback received through a user interface upon presenting, in the user interface, said at least one mean observer score prediction. Doing so, prediction quality can be improved after re-training at least for the selection.

Method and apparatus for measuring video quality using at least one semi-supervised learning regressor for mean observer score prediction

The purpose of an objective video quality evaluation is to automatically assess the quality of video sequences in agreement with human quality perception. The invention addresses the effects of the introduction of a temporal dimension by focusing on the temporal evolutions of spatial distortions, since it has been found that a spatial quality variation cannot be evaluated by simple subtraction of the spatial quality of neighboring frames. An improved method for estimating perceived video quality comprises steps of calculating a first similarity map between adjacent frames of a current sequence, calculating a second similarity map between the corresponding reference frames, and calculating (smg) a third similarity map, which provides a numerical quality value.

Method for measuring video quality using a reference, and apparatus for measuring video quality using a reference

A method for processing an image divided into blocks of pixels is disclosed. The method comprises the steps of:

Method of processing an image and corresponding device

A method and apparatus for measuring the quality of a video sequence, which includes a plurality of frames, among which one or more consecutive frames are lost, wherein during the displaying of the video sequence, the one or more lost frames are substituted by an immediate preceding frame in the video sequence during a period from the displaying of the immediate preceding frame to that of an immediate subsequent frame of the one or more lost frames, the method includes measuring the quality of the video sequence as a function of a first parameter (DF) relating to the stability of the immediate preceding frame during the period, a second parameter (DD) relating to the continuity between the immediate preceding frame and the immediate subsequent frame, and a third parameter (DP) relating to the coherent 20 motions of the video sequence.

Growing community of inventors

Beijing, China

Xiaodong Gu