3D object tracking using mean-shift and similarity-based aspect-graph modeling

The mean shift algorithm is a popular method in the field of 2D object tracking due to its simplicity and robustness over slight variations of lighting condition, scale and view-point over time. However, the appearance of 3D object might have distinctive variations for different viewpoints over time. In this work, a novel method for tracking 3D objects using mean-shift algorithm and a 3D object database is proposed to achieve a more precise tracking. A 3D object database using similarity-based aspect-graph is built from 2D images sampled at random intervals from the viewing sphere. Contour and color features of each 2D image are used for modeling the 3D object database. To conduct tracking, a suitable object model is selected from the database and the mean-shift tracking is applied to find the local minima of a similarity measure between the color histograms of the object model and the target image. The effectiveness of the proposed method is demonstrated by experiments with objects rotating and translating in space.