公开(公告)号：US20240296623A1

公开(公告)日：2024-09-05

申请号：US18169825

申请日：2023-02-15

Applicant: Nvidia Corporation

Inventor： Jiahui Huang ,  Francis Williams ,  Zan Gojcic ,  Matan Atzmon ,  Or Litany ,  Sanja Fidler

IPC: G06T17/20 ,  G06T15/08

CPC classification number: G06T17/20 ,  G06T15/08 ,  G06T2210/56

Abstract: Approaches presented herein provide for the reconstruction of implicit multi-dimensional shapes. In one embodiment, oriented point cloud data representative of an object can be obtained using a physical scanning process. The point cloud data can be provided as input to a trained density model that can infer density functions for various points. The points can be mapped to a voxel hierarchy, allowing density functions to be determined for those voxels at the various levels that are associated with at least one point of the input point cloud. Contribution weights can be determined for the various density functions for the sparse voxel hierarchy, and the weighted density functions combined to obtain a density field. The density field can be evaluated to generate a geometric mesh where points having a zero, or near-zero, value are determined to contribute to the surface of the object.

公开(公告)号：US20250131685A1

公开(公告)日：2025-04-24

申请号：US18674668

申请日：2024-05-24

Applicant: NVIDIA Corporation

Inventor： Sanja FIDLER ,  Matan Atzmon ,  Jiahui Huang ,  Or Litany ,  Francis Williams

IPC: G06V10/44 ,  G06V10/82

Abstract: In various examples, a technique for modeling equivariance in point neural networks includes generating, via execution of one or more layers included in a neural network, a set of features associated with a first partition prediction for a plurality of points included in a scene. The technique also includes applying, to the set of features, one or more transformations included in a frame associated with the plurality of points to generate a set of equivariant features. The technique further includes generating a second partition prediction for the plurality of points based at least on the set of equivariant features, and causing an object recognition result associated with the plurality of points to be generated based at least on the second partition prediction.

公开(公告)号：US20240005604A1

公开(公告)日：2024-01-04

申请号：US18320716

申请日：2023-05-19

Applicant: Nvidia Corporation

Inventor： Karsten Julian Kreis ,  Xiaohui Zeng ,  Arash Vahdat ,  Francis Williams ,  Zan Gojcic ,  Or Litany ,  Sanja Fidler

IPC: G06T17/20 ,  G06V10/44

CPC classification number: G06T17/20 ,  G06V10/44

Abstract: Approaches presented herein provide for the unconditional generation of novel three dimensional (3D) object shape representations, such as point clouds or meshes. In at least one embodiment, a first denoising diffusion model (DDM) can be trained to synthesize a 1D shape latent from Gaussian noise, and a second DDM can be trained to generate a set of latent points conditioned on this 1D shape latent. The shape latent and set of latent points can be provided to a decoder to generate a 3D point cloud representative of a random object from among the object classes on which the models were trained. A surface reconstruction process may be used to generate a surface mesh from this generated point cloud. Such an approach can scale to complex and/or multimodal distributions, and can be highly flexible as it can be adapted to various tasks such as multimodal voxel- or text-guided synthesis.