公开(公告)号：US11151734B2

公开(公告)日：2021-10-19

申请号：US16568885

申请日：2019-09-12

申请人： Lucas Pagé-Caccia ,  Joelle Pineau ,  Elmira Amirloo Abolfathi

发明人： Lucas Pagé-Caccia ,  Joelle Pineau ,  Elmira Amirloo Abolfathi

IPC分类号： G06T7/00 ,  G06T7/521 ,  G06N3/08 ,  G06N3/04 ,  G06N20/20 ,  G06N20/10

摘要： Methods and systems for generating synthetic point cloud data are described. Projected 2D data grid is generated by projecting a 3D point cloud into a 2D grid, with rotation equivariance. A generative model is learned using the projected 2D data grid, wherein the generative model is implemented using flex-convolution and transpose flex convolution operations, for example in a generative adversarial network. The learned generative model is used to generate synthetic point clouds.

公开(公告)号：US20200090357A1

公开(公告)日：2020-03-19

申请号：US16568885

申请日：2019-09-12

申请人： Lucas PAGÉ-CACCIA ,  Joelle PINEAU ,  Elmira AMIRLOO ABOLFATHI

发明人： Lucas PAGÉ-CACCIA ,  Joelle PINEAU ,  Elmira AMIRLOO ABOLFATHI

IPC分类号： G06T7/521 ,  G06N3/08 ,  G06N3/04 ,  G06N20/10 ,  G06N20/20

摘要： Methods and systems for generating synthetic point cloud data are described. Projected 2D data grid is generated by projecting a 3D point cloud into a 2D grid, with rotation equivariance. A generative model is learned using the projected 2D data grid, wherein the generative model is implemented using flex-convolution and transpose flex convolution operations, for example in a generative adversarial network. The learned generative model is used to generate synthetic point clouds.