Skip to content

Latest commit

 

History

History
40 lines (28 loc) · 1.46 KB

spherical_image_processing_4_VO.md

File metadata and controls

40 lines (28 loc) · 1.46 KB

Spherical Image Processing for Accurate Visual Odometry with Omnidirectional Cameras

Author: Hicham Hadj-Abdelkader

Year: 2008

Notes:

  • perspective image tools are unable to handle geometrical distortion
  • images are mapped onto a unit sphere and treated in the spherical spectral domain

Intro:

  • feature based: extracting features + matching and pose estimation (low accuracy but wide baselinge)
  • template based: using the whole images (high accuracy but small baseline)
  • Here uses feature based method to initialize template based method
  • focus on feature based using Harris point detector and SIFT descriptor

Background

  • mapping of the image on the unit sphere:
    • sampling a equi angular grid on the unit sphere
    • spherical point mapped on the image plane with camera model
    • interpolation in the omnidirectionnal plane
  • spherical spectral domain: $I$ can be expanded into a linear summation of spherical harmonic function
  • convolution is defined in $SO(3)$ for spherical images

Spherical image processing

  • gaussian filtering is usually performed before feature extraction => spherical gaussian filter
  • Definition of spherical derivatives (w.r.t. $\theta$ and $\phi$)

Feature based approach

  • use spherical derivatives for Harris corners
  • comparison using the repeatability rate
  • SIFT descriptors are built on a circular region in the spherical image
  • Odometry using only $E$ estimation with RANSAC
  • => improvement vs classical methods