Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Learning and propagating visual attributes

A point collection, spatial distribution technology, applied in the field of learning and dissemination of visual attributes, can solve problems such as restricted reasoning or dissemination

Pending Publication Date: 2022-03-29
NVIDIA CORP
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

As a result, these techniques may not be suitable for use with point clouds and / or other unstructured data that lack fixed topology and / or ordering, which limits the performance of class labels, colors, and / or other types of attributes on unstructured data. ability to reason or communicate

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Learning and propagating visual attributes
  • Learning and propagating visual attributes
  • Learning and propagating visual attributes

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0014] In the following description, numerous specific details are set forth in order to provide a more thorough understanding of various embodiments. It will be apparent, however, to one skilled in the art that the inventive concepts may be practiced without one or more of these specific details.

[0015] general overview

[0016] Computer vision and / or image processing tasks often involve determining or refining visual properties of objects or scenes. For example, an image may be analyzed or processed to identify an animal, person, car, road, building, body part, clothing, furniture, or other object in the image; determine the pixel location of the object within the image; the resolution of the above image; or add or enhance color to the image. Recognizing this visual property of images is important in many real-world applications, including autonomous vehicles, medical imaging, and robotic automation of manufacturing facilities and other processes.

[0017] These types o...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

Learning and propagating visual attributes are disclosed. One embodiment of the present invention proposes a technique for performing spatial propagation. The technique includes generating a first directed acyclic graph (DAG) by connecting spatially adjacent points included in a set of unstructured points via directed edges in a first direction. The technique further includes applying a first set of neural network layers to one or more images associated with the set of unstructured points to generate (i) a set of features for the set of unstructured points and (ii) a set of pairs of relevancy between spatially adjacent points connected by directed edges. The technique further includes generating a set of tags for the set of unstructured points by propagating the set of features across the first DAG based on the set of pairs of relevancy.

Description

technical field [0001] Embodiments of the present disclosure relate generally to spatial propagation, and more specifically, to learning and propagating visual attributes in arbitrary structured data. Background technique [0002] In computer vision, correlation matrices include general matrices that characterize the similarity or proximity of two points in space. For example, a correlation matrix may store a set of weights for pairs of pixels in an image. Each weight may represent a semantic similarity or proximity between corresponding pairs of pixels; a weight closer to 0 indicates a lack of semantic similarity or proximity, while a weight closer to 1 indicates a high semantic similarity or proximity. [0003] The correlation matrix can then be used to perform and / or enhance various image processing or computer vision tasks. For example, pairwise correlations between pixels in an image can be used to perform or refine semantic segmentation, coloring, filtering, clusteri...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06V10/74G06V10/56G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06N3/084G06N3/045G06F18/22G06V10/761G06V10/765G06V10/774G06N3/048G06F18/29G06F18/214G06F18/2163G06T17/00G06V10/82
Inventor J·考茨刘思飞S·德·梅洛V·扬帕尼
Owner NVIDIA CORP
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com