抄録
I-013
Adaptive Edge Detection and Tracking for Robust Model-Based Camera Tracking
○朴　漢薰・三ッ峰秀樹・藤井真人（NHK放送技研）
In adverse real environment, edges on a 3D scene model usually have multiple candidate correspondences (or hypotheses) and there is little information on which one is the true hypothesis. This ambiguity makes model-based camera tracking unstable and inaccurate. Therefore, this paper proposes an adaptive edge detection and tracking method that models the gradients of true hypotheses as a mixture of Gaussian distributions and selectively eliminates false hypotheses. The method reduced the pose error and jitter of a testbed model-based camera tracking system by 10% and 6%, respectively.