公开(公告)号：US11995883B2

公开(公告)日：2024-05-28

申请号：US18163482

申请日：2023-02-02

Applicant: Nvidia Corporation

Inventor： Aayush Prakash ,  Shoubhik Debnath ,  Jean-Francois Lafleche ,  Eric Cameracci ,  Gavriel State ,  Marc Teva Law

IPC: G06V10/82 ,  G06F18/10 ,  G06F18/20 ,  G06F18/24 ,  G06V10/764 ,  G06V10/84 ,  G06V20/00 ,  G06V20/56 ,  G06V20/70

CPC classification number: G06V10/82 ,  G06F18/10 ,  G06F18/24 ,  G06F18/29 ,  G06V10/764 ,  G06V10/84 ,  G06V20/00 ,  G06V20/70 ,  G06V20/56

Abstract: Approaches are presented for training and using scene graph generators for transfer learning. A scene graph generation technique can decompose a domain gap into individual types of discrepancies, such as may relate to appearance, label, and prediction discrepancies. These discrepancies can be reduced, at least in part, by aligning the corresponding latent and output distributions using one or more gradient reversal layers (GRLs). Label discrepancies can be addressed using self-pseudo-statistics collected from target data. Pseudo statistic-based self-learning and adversarial techniques can be used to manage these discrepancies without the need for costly supervision from a real-world dataset.

公开(公告)号：US20230177826A1

公开(公告)日：2023-06-08

申请号：US18163482

申请日：2023-02-02

Applicant: Nvidia Corporation

Inventor： Aayush Prakash ,  Shoubhik Debnath ,  Jean-Francois Lafleche ,  Eric Cameracci ,  Gavriel State ,  Marc Teva Law

IPC: G06V10/82 ,  G06V20/70 ,  G06F18/10 ,  G06F18/24 ,  G06F18/20 ,  G06V10/764 ,  G06V10/84 ,  G06V20/00

CPC classification number: G06V10/82 ,  G06V20/70 ,  G06F18/10 ,  G06F18/24 ,  G06F18/29 ,  G06V10/764 ,  G06V10/84 ,  G06V20/00 ,  G06V20/56

Abstract: Approaches are presented for training and using scene graph generators for transfer learning. A scene graph generation technique can decompose a domain gap into individual types of discrepancies, such as may relate to appearance, label, and prediction discrepancies. These discrepancies can be reduced, at least in part, by aligning the corresponding latent and output distributions using one or more gradient reversal layers (GRLs). Label discrepancies can be addressed using self-pseudo-statistics collected from target data. Pseudo statistic-based self-learning and adversarial techniques can be used to manage these discrepancies without the need for costly supervision from a real-world dataset.

公开(公告)号：US11182649B2

公开(公告)日：2021-11-23

申请号：US17119971

申请日：2020-12-11

Applicant: NVIDIA Corporation

Inventor： Jonathan Tremblay ,  Aayush Prakash ,  Mark A. Brophy ,  Varun Jampani ,  Cem Anil ,  Stanley Thomas Birchfield ,  Thang Hong To ,  David Jesus Acuna Marrero

IPC: G06K9/62 ,  G06T15/04 ,  G06T15/50 ,  G06T15/20 ,  G06N3/04 ,  G06N3/08

Abstract: Training deep neural networks requires a large amount of labeled training data. Conventionally, labeled training data is generated by gathering real images that are manually labelled which is very time-consuming. Instead of manually labelling a training dataset, domain randomization technique is used generate training data that is automatically labeled. The generated training data may be used to train neural networks for object detection and segmentation (labelling) tasks. In an embodiment, the generated training data includes synthetic input images generated by rendering three-dimensional (3D) objects of interest in a 3D scene. In an embodiment, the generated training data includes synthetic input images generated by rendering 3D objects of interest on a 2D background image. The 3D objects of interest are objects that a neural network is trained to detect and/or label.