Real-Time Camera Parameter Estimation from Images for a Wearable Vision System

In this paper, we propose a method of estimating camera parameters for annotating 3D virtual objects with a wearable vision system. The proposed method tracks natural feature points using a doubled color histogram matching and template matching. The model-view matrix calculation method consists of three algorithms for PnP problems and uses each algorithm selectively. The experimental system using the proposed method works at 5 to 10 frames per second.

• Type: Conference
• Year: 2000
• Where: MVA
• Authors: Takashi Okuma, Takeshi Kurata, Katsuhiko Sakaue
• Keywords: Color Histogram Matching, Matrix Calculation Method, Method Tracks, MVA 2000, MVA 2007