Automatic camera pose initialization, using scale, rotation and luminance invariant natural feature tracking
Year: 2008
Type: article
Abstract: The solution to the camera registration and tracking problem serves Augmented Reality, in order to provide an\nenhancement to the user’s cognitive perception of the real world and his/her situational awareness. By analyzing\nthe five most representative tracking and feature detection techniques, we have concluded that the Camera Pose\nInitialization (CPI) problem, a relevant sub-problem in the overall camera tracking problem, is still far from being\nsolved using straightforward and non-intrusive methods. The assessed techniques often use user inputs (i.e.\nmouse clicking) or auxiliary artifacts (i.e. fiducial markers) to solve the CPI problem. This paper presents a novel\napproach to real-time scale, rotation and luminance invariant natural feature tracking, in order to solve the CPI\nproblem using totally automatic procedures. The technique is applicable for the case of planar objects with arbitrary\ntopologies and natural textures, and can be used in Augmented Reality. We also present a heuristic method\nfor feature clustering, which has revealed to be efficient and reliable. The presented work uses this novel feature\ndetection technique as a baseline for a real-time and robust planar texture tracking algorithm, which combines\noptical flow, backprojection and template matching techniques. The paper presents also performance and precision\nresults of the proposed technique.
Source: Repositório do ISCTE-IUL
Authors: Rafael Bastos, Miguel Sales Dias
Institution: Microsoft (United States)
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2400875305