公开(公告)号：US20240282055A1

公开(公告)日：2024-08-22

申请号：US18170487

申请日：2023-02-16

Applicant: QUALCOMM Incorporated

Inventor： Shraman Biswas ,  Joseph Daniel GARVEY ,  Bojan VRCELJ

IPC: G06T17/20 ,  G06T9/00

CPC classification number: G06T17/20 ,  G06T9/00

Abstract: This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for a mesh GPU codec for real-time streaming. A graphics processor may obtain an indication of a compressed representation of a topology associated with at least one generalized triangle strip associated with a subset of triangles in a set of triangles. The graphics processor may obtain an intermediate representation of the topology associated with the at least one generalized triangle strip by a decompression of the compressed representation. The graphics processor may obtain a sequence of indices based on the intermediate representation of the topology associated with the at least one generalized triangle strip. The graphics processor may compute at least one set of offsets based on the intermediate representation. The graphics processor may reconstruct the set of triangles based on the sequence of indices and the at least one set of offsets.

公开(公告)号：US20240273829A1

公开(公告)日：2024-08-15

申请号：US18429149

申请日：2024-01-31

Applicant: QUALCOMM Incorporated

Inventor： Thomas STOCKHAMMER ,  Imed BOUAZIZI ,  Bojan VRCELJ ,  Prashanth Haridas HANDE ,  Liangping MA

IPC: G06T19/00 ,  G06T7/70 ,  H04L67/04

CPC classification number: G06T19/006 ,  G06T7/70 ,  H04L67/04 ,  G06T2207/10021

Abstract: Extended reality (XR) management systems and techniques are described. In some examples, an XR management system receives sensor data from an XR interface device having at least one sensor. The XR management system generates, based on the receipt of the sensor data from the XR interface device, processing instructions for an XR processing device to process the sensor data to generate XR content. The XR management system sends the sensor data and the processing instructions to the XR processing device. The XR management system receives the XR content from the XR processing device. The XR management system generates layer content. The XR management system sends the XR content and the layer content to the XR interface device to cause the XR interface device to output the XR content and the layer content in a layered arrangement.

公开(公告)号：US20240146428A1

公开(公告)日：2024-05-02

申请号：US18051199

申请日：2022-10-31

Applicant: QUALCOMM Incorporated

Inventor： Surbhit SINHA ,  Peerapol TINNAKORNSRISUPHAP ,  Bojan VRCELJ

IPC: H04B17/373 ,  H04L67/131

CPC classification number: H04B17/373 ,  H04L67/131

Abstract: Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a wireless communication device may predict user behavior in an extended reality (XR) application on a user equipment (UE) based at least in part on current content presented by the XR application. The wireless communication device may predict channel information for a communication link of the UE based at least in part on the predicted user behavior. The wireless communication device may adjust one or more parameters of the communication link based at least in part on the predicted channel information. Numerous other aspects are described.

公开(公告)号：US20240104794A1

公开(公告)日：2024-03-28

申请号：US18168532

申请日：2023-02-13

Applicant: QUALCOMM Incorporated

Inventor： Joseph Daniel GARVEY ,  Bojan VRCELJ

IPC: G06T11/00 ,  G06T1/60 ,  G06T7/00 ,  G06T7/49

CPC classification number: G06T11/001 ,  G06T1/60 ,  G06T7/0002 ,  G06T7/49 ,  G06T2207/10016 ,  G06T2207/30168

Abstract: This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for PVS over UDP for split rendering. A graphics processor may obtain, a PVS and a set of video packets. The graphics processor may determine, if a data loss associated with the PVS or the set of video packets is detected, one or more first visible primitives in the set of visible primitives that are currently affected by the data loss associated with the PVS or the set of video packets. The graphics processor may identify concealment information for the one or more first visible primitives in the set of visible primitives based on the one or more first visible primitives in a previous frame. The graphics processor may update information associated with the one or more first visible primitives based on the concealment information for the one or more first visible primitives.

公开(公告)号：US20210312704A1

公开(公告)日：2021-10-07

申请号：US17087252

申请日：2020-11-02

Applicant: QUALCOMM Incorporated

Inventor： Joseph Daniel GARVEY ,  Bojan VRCELJ

IPC: G06T15/60 ,  G06T19/00 ,  G06T7/60 ,  G06T15/80 ,  H04N19/46 ,  H04N21/236

Abstract: For a scene comprising real-world objects and generated (CG) objects, the scene characterized by a scene geometry of a first time, one or more processors can generate, based on the scene geometry, shadow information for each of one or more shadows cast by a CG object of the scene on one or more real-world objects of the scene. The one or more processors can rendering and display a frame of the scene as a function of the shadow information and a scene geometry at a time of the rendering later than the first time. In some examples, the shadow information includes shadow factors and identifiers. Each shadow factor can describe a light attenuation effect of a cast shadow on a real world object surface and can be incorporated into a shading atlas of the scene. Each identifier identifies a real-world object surface of the scene affected by a shadow factor.

公开(公告)号：US20150095360A1

公开(公告)日：2015-04-02

申请号：US14492976

申请日：2014-09-22

Applicant: QUALCOMM Incorporated

Inventor： Bojan VRCELJ ,  Raghuraman KRISHNAMOORTHI ,  Qi PAN ,  Wonwoo LEE ,  Emilio MAGGIO

IPC: G06F17/30

CPC classification number: G06F16/24 ,  G06F16/444 ,  G06F16/583 ,  G06F16/5854 ,  G06K9/6211 ,  G06K9/6274

Abstract: A method of building a database for an object recognition system includes acquiring several multi-view images of a target object and then extracting a first set of features from the images. One of these extracted features is then selected and a second set of features is determined based on which of the first set of features include both, descriptors that match and keypoint locations that are proximate to the selected feature. If a repeatability of the selected feature is greater than a repeatability threshold and if a discriminability is greater than a discriminability threshold, then at least one derived feature is stored to the database, where the derived single feature is representative of the second set of features.

Abstract translation: 构建用于对象识别系统的数据库的方法包括获取目标对象的多个多视图图像，然后从图像中提取第一组特征。 然后选择这些提取的特征中的一个，并且基于第一组特征中的哪一个包括与所选择的特征相邻的描述符和匹配关键点的位置来确定第二组特征。 如果所选特征的可重复性大于可重复性阈值，并且如果可辨别性大于可识别阈值，则至少一个导出特征被存储到数据库，其中所导出的单个特征代表第二组特征。

公开(公告)号：US20140270344A1

公开(公告)日：2014-09-18

申请号：US13796444

申请日：2013-03-12

Applicant: QUALCOMM INCORPORATED

Inventor： Raghuraman KRISHNAMOORTHI ,  Gary G. McGRATH ,  Bojan VRCELJ

IPC: G06K9/46

CPC classification number: G06K9/46 ,  G06K9/6211 ,  G06K9/6857

Abstract: In one example, a method for exiting an object detection pipeline includes determining, while in the object detection pipeline, a number of features within a first tile of an image, wherein the image consists of a plurality of tiles, performing a matching procedure using at least a subset of the features within the first tile if the number of features within the first tile meets a threshold value, exiting the object detection pipeline if a result of the matching procedure indicates an object is recognized in the image, and presenting the result of the matching procedure.

Abstract translation: 在一个示例中，用于退出对象检测流水线的方法包括在对象检测流水线中确定图像的第一瓦片内的多个特征，其中所述图像由多个瓦片组成，其中使用 如果所述第一瓦片内的特征数目满足阈值，则在所述第一瓦片内的所述特征的至少一部分，如果所述匹配过程的结果指示对象在所述图像中被识别，则退出所述对象检测流水线， 匹配程序。