公开(公告)号：US20210233326A1

公开(公告)日：2021-07-29

申请号：US15734486

申请日：2018-09-28

Applicant: Intel Corporation

Inventor： Wenlong Li ,  Yuri Shpalensky ,  Alex Gendelman ,  Maria Bortman ,  Asaf Shiloni

IPC: G06T19/20 ,  G06T7/73 ,  G06T17/00

Abstract: Methods and apparatus to generate photo-realistic three-dimensional models of a photographed environment are disclosed. An apparatus includes an object position calculator to determine a three-dimensional (3D) position of an object detected within a first image of an environment and within a second image of the environment. The apparatus further includes a 3D model generator to generate a 3D model of the environment based on the first image and the second image. The apparatus also includes a model integrity analyzer to detect a difference between the 3D position of the object and the 3D model. The 3D model generator automatically modifies the 3D model based on the difference in response to the difference satisfying a confidence threshold.

公开(公告)号：US10819968B2

公开(公告)日：2020-10-27

申请号：US16050285

申请日：2018-07-31

Applicant: Intel Corporation

Inventor： Ben Raziel ,  Asaf Shiloni ,  Gilad Bauman ,  Jill Boyce ,  Tzach Ashkenazi ,  Ilan Beer ,  Eytan Kats

IPC: H04N13/156 ,  H04N13/332 ,  G06N3/08 ,  G06N5/04 ,  H04N13/161 ,  G06N20/00

Abstract: Embodiments are generally directed to neural network based patch blending for immersive video. An embodiment of a system includes one or more processor cores; a memory to store data for images in immersive video; and a neural network training framework. The neural network training framework is to generate a trained neural network for blending of a view of an object with patches generated for one or more detected occlusions for the object.

公开(公告)号：US11551428B2

公开(公告)日：2023-01-10

申请号：US15734486

申请日：2018-09-28

Applicant: Intel Corporation

Inventor： Wenlong Li ,  Yuri Shpalensky ,  Alex Gendelman ,  Maria Bortman ,  Asaf Shiloni

IPC: G06T19/20 ,  G06T17/00 ,  G06T7/70 ,  G06T7/73

Abstract: Methods and apparatus to generate photo-realistic three-dimensional models of a photographed environment are disclosed. An apparatus includes an object position calculator to determine a three-dimensional (3D) position of an object detected within a first image of an environment and within a second image of the environment. The apparatus further includes a 3D model generator to generate a 3D model of the environment based on the first image and the second image. The apparatus also includes a model integrity analyzer to detect a difference between the 3D position of the object and the 3D model. The 3D model generator automatically modifies the 3D model based on the difference in response to the difference satisfying a confidence threshold.

公开(公告)号：US20210281887A1

公开(公告)日：2021-09-09

申请号：US17253182

申请日：2018-11-12

Applicant: Intel Corporation

Inventor： Xiaofeng Tong ,  Eli Turiel ,  Qiang Li ,  Wenlong Li ,  Yikai Fang ,  Doron Houminer ,  Asaf Shiloni

IPC: H04N21/218 ,  H04N21/6587 ,  A63B24/00 ,  H04N5/232

Abstract: Methods, systems and apparatuses may provide for technology that identifies a player captured in a multi-camera video feed of a game that involves the identified player and estimates a first field of view from the perspective of the identified player for a selected from of the multi-camera video feed. Additionally, the technology automatically generates, based on the first field of view, a camera path for a replay of the selected frame from the perspective of the identified player. In one example, the technology also determines a trajectory of a projectile captured in the multi-camera video feed, estimates, based on the trajectory, a second field of view from the perspective of the projectile, and automatically generates, based on the second field of view, a replay of one or more selected frames of the multi-camera video feed from the perspective of the projectile.

公开(公告)号：US20200045289A1

公开(公告)日：2020-02-06

申请号：US16050285

申请日：2018-07-31

Applicant: Intel Corporation

Inventor： Ben Raziel ,  Asaf Shiloni ,  Gilad Bauman ,  Jill Boyce ,  Tzach Ashkenazi ,  Ilan Beer ,  Eytan Kats

IPC: H04N13/156 ,  H04N13/332 ,  H04N13/161 ,  G06N3/08 ,  G06N5/04 ,  G06N99/00

Abstract: Embodiments are generally directed to neural network based patch blending for immersive video. An embodiment of a system includes one or more processor cores; a memory to store data for images in immersive video; and a neural network training framework. The neural network training framework is to generate a trained neural network for blending of a view of an object with patches generated for one or more detected occlusions for the object.