Assessment of Camera Pose Estimation Using Geo-Located Images from Simultaneous Localization and Mapping

Abstract

This research proposes a method for enabling low-cost camera localization using geo-located images generated with factorgraph-based Simultaneous Localization And Mapping (SLAM). The SLAM results are paired with panoramic image data to generate geo-located images, which can be used to locate and orient low-cost cameras. This study determines the efficacy of using a spherical camera and LIDAR sensor to enable localization for a wide range of cameras with low size, weight, power, and cost. This includes determining the accuracy of SLAM when geo-referencing images, along with introducing a promising method for extracting range measurements from monocular images of known features.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Feb 26, 2019
Accession Number
AD1074427

Entities

People

  • David W. Beargie

Organizations

  • Air Force Institute of Technology

Tags

Communities of Interest

  • Autonomy
  • Energy and Power Technologies
  • Sensors
  • Space

DTIC Thesaurus Topics

  • Air Force
  • Autonomous Navigation
  • Computers
  • Coordinate Systems
  • Databases
  • Department Of Defense
  • Detection
  • Detectors
  • Geometry
  • Image Processing
  • Information Science
  • Kalman Filters
  • Measurement
  • Mobile Phones
  • Operating Systems
  • Simultaneous Localization And Mapping
  • United States Government

Fields of Study

  • Computer science

Readers

  • Computer Vision.