公开(公告)号：US20250086090A1

公开(公告)日：2025-03-13

申请号：US18244206

申请日：2023-09-08

Applicant: NVIDIA Corporation

Inventor： Durgadoss R ,  Hariharan Sandanagobalane ,  Pradeep Kumar ,  Khaled Abdul Karim Mohammed ,  Vatsa Santhanam

IPC: G06F11/34 ,  G06F8/41

Abstract: Apparatuses, systems, and techniques to perform variable value profiling. In at least one embodiment, variable value profiling is to be performed by one or more accelerators.

公开(公告)号：US20250086036A1

公开(公告)日：2025-03-13

申请号：US18463451

申请日：2023-09-08

Applicant: NVIDIA Corporation

Inventor： Mohamed Saad Gaber Abdelhameed ,  Vito Magnanimo

IPC: G06F11/00 ,  B60W50/02

Abstract: In various examples, confirming or evaluating availability requirements as part of a safety analysis for autonomous and semi-autonomous systems and applications is described herein. For instance, systems and methods are disclosed that may add availability requirements to one or more safety analysis techniques, such as failure mode and effects analysis (FMEA), critical path analysis (CPA), and/or any other safety analysis technique. In some examples, an availability requirement may indicate whether a failure mode associated with a detected failure will impact an availability of an element and/or one or more measures that may be taken to avoid impacting the availability of the element. Systems and methods may then analyze a machine using these updated safety analysis techniques to detect failure modes in a system, causes of the failure modes, and effects of the failure modes, as well as determine whether these failure modes will impact the availability of the system.

公开(公告)号：US12248788B2

公开(公告)日：2025-03-11

申请号：US17691690

申请日：2022-03-10

Applicant: NVIDIA Corporation

Inventor： Prakash Bangalore Prabhakar ,  Gentaro Hirota ,  Ronny Krashinsky ,  Ze Long ,  Brian Pharris ,  Rajballav Dash ,  Jeff Tuckey ,  Jerome F. Duluk, Jr. ,  Lacky Shah ,  Luke Durant ,  Jack Choquette ,  Eric Werness ,  Naman Govil ,  Manan Patel ,  Shayani Deb ,  Sandeep Navada ,  John Edmondson ,  Greg Palmer ,  Wish Gandhi ,  Ravi Manyam ,  Apoorv Parle ,  Olivier Giroux ,  Shirish Gadre ,  Steve Heinrich

IPC: G06F9/30 ,  G06F9/38 ,  G06F9/52 ,  G06F9/54 ,  G06F13/16 ,  G06T1/60

Abstract: Distributed shared memory (DSMEM) comprises blocks of memory that are distributed or scattered across a processor (such as a GPU). Threads executing on a processing core local to one memory block are able to access a memory block local to a different processing core. In one embodiment, shared access to these DSMEM allocations distributed across a collection of processing cores is implemented by communications between the processing cores. Such distributed shared memory provides very low latency memory access for processing cores located in proximity to the memory blocks, and also provides a way for more distant processing cores to also access the memory blocks in a manner and using interconnects that do not interfere with the processing cores' access to main or global memory such as hacked by an L2 cache. Such distributed shared memory supports cooperative parallelism and strong scaling across multiple processing cores by permitting data sharing and communications previously possible only within the same processing core.

公开(公告)号：US20250081349A1

公开(公告)日：2025-03-06

申请号：US18242290

申请日：2023-09-05

Applicant: NVIDIA CORPORATION

Inventor： Elad Mentovich ,  Dongji Xie ,  Ron Chao ,  Ryan Albright

IPC: H05K1/18 ,  H05K1/02 ,  H05K3/12 ,  H05K3/28

Abstract: A secure electronic component assembly is described herein for ensuring the physical integrity of an integrated circuit (IC). The secure electronic component assembly may comprise a printed circuit board (PCB), an integrated circuit (IC) mounted on the PCB, and an underfill material disposed between the IC and the PCB. The underfill material comprises a detection agent that is configured to change a state of the IC in response to exposure to an external environment, wherein the change in the state of the IC is indicative of a tamper condition.

公开(公告)号：US20250080863A1

公开(公告)日：2025-03-06

申请号：US18242395

申请日：2023-09-05

Applicant: NVIDIA Corporation

Inventor： Chengwu Cui ,  Sean Midthun Pieper ,  Du-Yong Ng ,  Douglas John Taylor

IPC: H04N23/84 ,  G06T7/00 ,  G06T7/90 ,  G06T11/00 ,  G06V20/52

Abstract: Apparatuses, systems, and techniques for performing color correction are presented. In at least one embodiment, a color mapping model may be identified that maps colors, within a subspace of an input color space localized around a target color, to an adjusted color space and applied to an input image to adjust a value of one or more pixels of the input image that fall within the subspace. In at least one embodiment, a color mapping model may be initialized that maps colors, within a subspace of an input color space localized around a target color, to an adjusted color space. At least one parameter of the color mapping model may be adjusted to reduce an amount of visible artifacts produced by the color mapping model.

公开(公告)号：US20250078458A1

公开(公告)日：2025-03-06

申请号：US18456984

申请日：2023-08-28

Applicant: NVIDIA Corporation

Inventor： Weiheng Chai ,  Anurag Singh ,  Yifang Xu

IPC: G06V10/764 ,  G06V10/44 ,  G06V10/774 ,  G06V10/776 ,  G06V10/778 ,  G06V20/40

Abstract: In various examples, data mining using group classifiers for autonomous and semi-autonomous systems and applications is described. For instance, systems and methods may train and use a classifier that is configured to determine whether data samples (e.g., images) represent objects associated with a group in various training and/or mining iterations. For instance, after each iteration, the classifier may become more accurate such that the classifier is able to better identify first data samples that are associated with the group (e.g., positive images) and/or second data samples that are not associated with the group (e.g., negative images). Additionally, the systems and methods may use another classifier to determine classifications associated with objects represented by at least the first data samples that are associated with the group. The systems and method may then store these first data samples and the classifications in one or more databases.

公开(公告)号：US20250078199A1

公开(公告)日：2025-03-06

申请号：US18239733

申请日：2023-08-29

Applicant: NVIDIA Corporation

Inventor： Wish GANDHI ,  Ze LONG ,  Kun FANG

IPC: G06T1/60

Abstract: A GPU can selectively confine software function execution and associated data storage resources to locally-connected processing/storage components, thereby minimizing latency and other overhead that would otherwise be needed to access more remote resources. The GPU can selectively permit other software function execution and associated data storage resources to range across non-locally-connected processing/storage components when more processing and/or storage resources are required.

公开(公告)号：US12245360B2

公开(公告)日：2025-03-04

申请号：US17879620

申请日：2022-08-02

Applicant: NVIDIA Corporation

Inventor： David Mohr ,  Tahir Cader ,  Elad Mentovich ,  Boaz Atias ,  Chong Tan

IPC: H05K1/02 ,  H05K7/20

Abstract: A power arrangement and cooling system for electronic devices includes a processing unit electrically interconnected and disposed to a first side of a printed circuit board and a power converter electrically interconnected and disposed on a second side of the printed circuit board. The power converter includes a thermally conductive material layer that transfers heat generated by the power converter during operation away from the power converter toward a heatsink that is disposed adjacent the second side of the printed circuit board. Heat generated by the processing unit is absorbed by a heatsink disposed adjacent the first side of the printed circuit board. Power for the processing unit is provided, by the power converter, through a thickness of the printed circuit board.

公开(公告)号：US20250069309A1

公开(公告)日：2025-02-27

申请号：US18454752

申请日：2023-08-23

Applicant: NVIDIA Corporation

Inventor： Ruthie D. Lyle

IPC: G06T13/40 ,  G06T19/00

Abstract: Apparatuses, systems, and techniques simulating physical properties of real-world objects. Data for a real-world object is collected. The collected data indicates one or more physical characteristics of the real-world object. A three-dimensional (3D) object is created based on some portion of the collected data. The 3D object has a format that is compatible with a 3D graphics platform. Physics simulation data associated with the 3D object moving within the 3D graphics platform is obtained. Image rendering data associated with the 3D object moving within the 3D graphics platform is stored with the computer system. The image rendering data is based at least in part on the obtained physics simulation data.

公开(公告)号：US20250069191A1

公开(公告)日：2025-02-27

申请号：US18452634

申请日：2023-08-21

Applicant: NVIDIA Corporation

Inventor： Iuri Frosio ,  Mayoore Selvarasa Jaiswal ,  Jan Kautz ,  Jianyuan Min

IPC: G06T5/50 ,  H04N23/743

Abstract: Systems and methods are disclosed related to synthetic bracketing for exposure correction. A deep learning based method and system produces a set of differently exposed images from a single input image. The images in the set may be combined to produce an output image with improved global and local exposure compared with the input image. An image encoder applies learned parameters to each input image to generate a set of image features including local exposure estimates for each of two or more regions of the input image and a low resolution latent representation of the input image. A decoder receives the local exposure estimates, the latent representation, and target enhancements that are processed to generate synthesized transformations. When applied to the input image, the synthesized transformations produce the set of transformed images. Each transformed image is a version of the input image synthesized to correspond to a respective target enhancement.