公开(公告)号：US11922575B2

公开(公告)日：2024-03-05

申请号：US17200400

申请日：2021-03-12

Applicant: A9.com, Inc.

Inventor： Himanshu Arora ,  Divyansh Agarwal ,  Arnab Dhua ,  Chun Kai Wang

IPC: G06T17/20 ,  G06T7/55

CPC classification number: G06T17/20 ,  G06T7/55 ,  G06T2200/08 ,  G06T2207/10028

Abstract: Approaches described and suggested herein relate to generating three-dimensional representations of objects to be used to render virtual reality and augmented reality effects on personal devices such as smartphones and personal computers, for example. An initial surface mesh of an object is obtained. A plurality of silhouette masks of the object taken from a plurality of viewpoints is also obtained. A plurality of depth maps are generated from the initial surface mesh. Specifically, the plurality of depth maps are taken from the same plurality of viewpoints from which the silhouette images are taken. A volume including the object is discretized into a plurality of voxels. Each voxel is then determined to be either inside the object or outside of the object based on the silhouette masks and the depth data. A final mesh is then generated from the voxels that are determined to be inside the object.

公开(公告)号：US20210201575A1

公开(公告)日：2021-07-01

申请号：US17200400

申请日：2021-03-12

Applicant: A9.com, Inc.

Inventor： Himanshu Arora ,  Divyansh Agarwal ,  Arnab Dhua ,  Chun Kai Wang

IPC: G06T17/20 ,  G06T7/55

Abstract: Approaches described and suggested herein relate to generating three-dimensional representations of objects to be used to render virtual reality and augmented reality effects on personal devices such as smartphones and personal computers, for example. An initial surface mesh of an object is obtained. A plurality of silhouette masks of the object taken from a plurality of viewpoints is also obtained. A plurality of depth maps are generated from the initial surface mesh. Specifically, the plurality of depth maps are taken from the same plurality of viewpoints from which the silhouette images are taken. A volume including the object is discretized into a plurality of voxels. Each voxel is then determined to be either inside the object or outside of the object based on the silhouette masks and the depth data. A final mesh is then generated from the voxels that are determined to be inside the object.

公开(公告)号：US20220221976A1

公开(公告)日：2022-07-14

申请号：US17148366

申请日：2021-01-13

Applicant: A9.com, Inc.

Inventor： Mukul Agarwal ,  Yu Lou ,  Anandram Sundar ,  Chun Kai Wang ,  Jack Mousseau ,  Kevin May ,  Xing Zhang ,  Yadikaer Yasheng ,  Geng Yan

IPC: G06F3/0481 ,  G06F3/0486 ,  G06F3/0484

Abstract: A virtual vertical surface in a three-dimensional space that represents a physical room may be detected. Responsive to a first user input gesture, movement of a virtual object within the three-dimensional space may be displayed. The movement may be to a first location in which a portion of the virtual object intersects a portion of the virtual vertical surface. A virtual vertical surface designator may be displayed corresponding to the virtual vertical surface based at least in part on the portion of the virtual object intersecting the portion of the virtual vertical surface. Upon determining that a second user input gesture meets or exceeds a movement threshold, movement of the three-dimensional object from the first location to a second location within the three-dimensional space may be displayed. The second location may appear beyond the virtual vertical surface.

公开(公告)号：US11138789B1

公开(公告)日：2021-10-05

申请号：US16452067

申请日：2019-06-25

Applicant: A9.com, Inc.

Inventor： Himanshu Arora ,  Divyansh Agarwal ,  Arnab Dhua ,  Chun Kai Wang

IPC: G06T17/20 ,  G06T17/00 ,  G06T7/521

Abstract: Approaches described and suggested herein relate to generating an enhanced point cloud representation of an objection and generating a surface mesh from the enhanced point cloud. The surface mesh can be used to render three-dimensional representations of objects on personal devices such as smartphones and personal computers, for example. Generating an enhanced point cloud of an object includes capturing a plurality of images of the object from a plurality of viewpoints about the object, generating an initial point cloud representation of the object from the plurality of images, generating a preliminary surface mesh from the point cloud using a Delauney-based meshing algorithm, and sampling points from the preliminary surface mesh. The sampled points are then added to the point cloud to form the enhanced point cloud. A final surface mesh can then be generated from the enhanced point cloud using a Poisson-based meshing algorithm.

公开(公告)号：US10991160B1

公开(公告)日：2021-04-27

申请号：US16452050

申请日：2019-06-25

Applicant: A9.com, Inc.

Inventor： Himanshu Arora ,  Divyansh Agarwal ,  Arnab Dhua ,  Chun Kai Wang

IPC: G06T17/20 ,  G06T17/00 ,  G06T7/55

Abstract: Approaches described and suggested herein relate to generating three-dimensional representations of objects to be used to render virtual reality and augmented reality effects on personal devices such as smartphones and personal computers, for example. An initial surface mesh of an object is obtained. A plurality of silhouette masks of the object taken from a plurality of viewpoints is also obtained. A plurality of depth maps are generated from the initial surface mesh. Specifically, the plurality of depth maps are taken from the same plurality of viewpoints from which the silhouette images are taken. A volume including the object is discretized into a plurality of voxels. Each voxel is then determined to be either inside the object or outside of the object based on the silhouette masks and the depth data. A final mesh is then generated from the voxels that are determined to be inside the object.