Unordered feature tracking made fast and easy

Pierre Moulon (1, 2, 3) , Pascal Monasse (1, 2, 3)

1 imagine [Marne-la-Vallée]
2 ENPC - École des Ponts ParisTech
3 LIGM - Laboratoire d'Informatique Gaspard-Monge

We present an efficient algorithm to fuse two-view correspondences into multi-view consistent tracks. The proposed method relies on the Union-Find algorithm to solve the fusion problem. It is very simple and has a lower computational complexity than other available methods. Our experiments show that it is faster and computes more tracks.

Keywords: tracking; Struture from Motion; computer vision

Domains: Computer Vision and Pattern Recognition [cs.CV]; Computer Science and Game Theory [cs.GT]

Pierre Moulon, Pascal Monasse. Unordered feature tracking made fast and easy. CVMP 2012, Dec 2012, London, United Kingdom. pp.1. ⟨hal-00769267⟩