公开(公告)号：US12014446B2

公开(公告)日：2024-06-18

申请号：US17409249

申请日：2021-08-23

Applicant: Google LLC

Inventor： Jing Yu Koh ,  Honglak Lee ,  Yinfei Yang ,  Jason Michael Baldridge ,  Peter James Anderson

IPC: G06T11/00 ,  G06F18/213 ,  G06N3/045 ,  G06N3/08 ,  G06T7/10 ,  G06T15/00 ,  G06T15/08

CPC classification number: G06T11/00 ,  G06F18/213 ,  G06N3/045 ,  G06N3/08 ,  G06T7/10 ,  G06T15/00 ,  G06T15/08 ,  G06T2207/10028 ,  G06T2207/20081

Abstract: A computing system for generating predicted images along a trajectory of unseen viewpoints. The system can obtain one or more spatial observations of an environment that may be captured from one or more previous camera poses. The system can generate a three-dimensional point cloud for the environment from the one or more spatial observations and the one or more previous camera poses. The system can project the three-dimensional point cloud into two-dimensional space to form one or more guidance spatial observations. The system can process the one or more guidance spatial observations with a machine-learned spatial observation prediction model to generate one or more predicted spatial observations. The system can process the one or more predicted spatial observations and image data with a machine-learned image prediction model to generate one or more predicted images from the target camera pose. The system can output the one or more predicted images.

公开(公告)号：US20230072293A1

公开(公告)日：2023-03-09

申请号：US17409249

申请日：2021-08-23

Applicant: Google LLC

Inventor： Jing Yu Koh ,  Honglak Lee ,  Yinfei Yang ,  Jason Michael Baldridge ,  Peter James Anderson

IPC: G06T11/00 ,  G06T15/00 ,  G06T15/08 ,  G06K9/62 ,  G06T7/10 ,  G06N3/08 ,  G06N3/04

Abstract: A computing system for generating predicted images along a trajectory of unseen viewpoints. The system can obtain one or more spatial observations of an environment that may be captured from one or more previous camera poses. The system can generate a three-dimensional point cloud for the environment from the one or more spatial observations and the one or more previous camera poses. The system can project the three-dimensional point cloud into two-dimensional space to form one or more guidance spatial observations. The system can process the one or more guidance spatial observations with a machine-learned spatial observation prediction model to generate one or more predicted spatial observations. The system can process the one or more predicted spatial observations and image data with a machine-learned image prediction model to generate one or more predicted images from the target camera pose. The system can output the one or more predicted images.