Method for initializing and solving the local geometry or surface normals of surfels using images in a parallelizable architecture

A surface normal and image technology, applied in the field of initializing and solving the local geometry or surface normal of the surface element using the image in the parallelizable framework, can solve the problem that the illumination change is not constant

Inactive Publication Date: 2015-07-08
QUALCOMM INC
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Due to computational limitations, the most traditional solutions to illumination changes are also not constant

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
  • Method for initializing and solving the local geometry or surface normals of surfels using images in a parallelizable architecture
  • Method for initializing and solving the local geometry or surface normals of surfels using images in a parallelizable architecture
  • Method for initializing and solving the local geometry or surface normals of surfels using images in a parallelizable architecture

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0020] figure 1 Composition diagram showing a computing device 100 that can track a camera in an unknown environment and / or can track an unknown object moving in front of the camera using estimated surface normals; the computing device 100 can be a desktop computer, laptop computer, mobile Telephone, cellular phone, cellular or other wireless communication device, personal communication system (PCS) device, personal navigation device (PND), personal information manager (PIM), personal digital assistant (PDA), tablet, or other similar device.

[0021] Computing device 100 may include a memory unit 102 and a processor 104 for performing operations of camera and / or object tracking in a captured video stream. In one embodiment, computing device 100 may include one or more video cameras 106a, b integrated into device 100 or connected to device 100 via wired or wireless peripheral connections. The video cameras 106a, b may capture video streams that will be used to perform camera a...

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

A system and method is described herein for solving for surface normals of objects in the scene observed in a video stream. The system and method may include sampling the video stream to generate a set of keyframes; generating hypothesis surface normals for a set of mappoints in each of the keyframes; warping patches of corresponding mappoints in a first keyframe to the viewpoint of a second keyframe with a warping matrix computed from each of the hypothesis surface normals; scoring warping errors between each hypothesis surface normal in the two keyframes; and discarding hypothesis surface normals with high warping errors between the first and second keyframes.

Description

technical field [0001] The subject matter disclosed herein relates generally to systems, methods, and apparatus for use in applications for mapping and tracking techniques, such as computer vision and augmented reality (AR) applications. In particular, systems and processes are used to compute surface normals by comparing hypothetical normals in a set of keyframes in a parallelizable framework. Background technique [0002] related background [0003] Applications such as gaming, computer vision, and augmented reality systems rely on accurate and efficient detection and tracking of objects. Tracking a camera in an unknown environment, or tracking an unknown object moving in front of the camera involves first building an appearance and structural representation of the environment or object. Tracking can then continue by matching image features from the incoming image with stored appearance representations that can be stored as feature patches as keyframes. [0004] To matc...

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
IPC IPC(8): G06T7/20
CPCG06T7/204G06T2207/10016G06K9/00711G06T7/248G06T17/20G06V20/40G06V20/64G06V20/653
Inventor 玛荷许·瑞玛屈安卓恩丹尼尔·瓦格纳穆拉利·拉马斯瓦米·查里
Owner QUALCOMM INC
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap