A novel genetic approach to epipolar geometry estimation

S. Chiesa, S. Taraglio

Research output: Contribution to conferencePaper

Abstract

A novel genetic approach to epipolar geometry estimation is presented. The individual genetically evolved in the algorithm encodes the rototranslation performed by a camera between two views of a scene, thus the fundamental matrix. The fitness function is designed to minimise the error between points and epipolar lines and to maximise the number of inliers. Results on synthetic and real images are reported and compared with data in literature.
Original languageEnglish
Publication statusPublished - 2010
Event2010 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2010 - , United States
Duration: 1 Jan 2010 → …

Conference

Conference2010 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2010
CountryUnited States
Period1/1/10 → …

    Fingerprint

All Science Journal Classification (ASJC) codes

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition

Cite this

Chiesa, S., & Taraglio, S. (2010). A novel genetic approach to epipolar geometry estimation. Paper presented at 2010 International Conference on Image Processing, Computer Vision, and Pattern Recognition, IPCV 2010, United States.