改良型MSEPFによる画像からの物体追跡
スポンサーリンク
概要
- 論文の詳細を見る
Recently, many approaches on applying particle filter to visual tracking problem have been proposed. However, it is hard to implement it to the real-time system because it requires a lot of computational and resources in order to achieve higher accuracy. As a method for reduce the computation time, Shan and coworkers proposed combining particle filter and Mean-Shift in order to keep the accuracy with small number of particles. In their approach, the state of each particle moves to the point in the window with the highest likelihood value. It is known that the accuracy of estimation depends on the size of the window, but the larger window size make the computation slower. In this paper, the authors propose a method for exploring the highest likelihood more quickly by means of random sampling. Moreover the authors propose multiple prediction models and new likelihood function that defines likelihood in terms of not only color cue but also motion cue.The effectiveness of the proposed method is evaluated by real image sequence experiments.
- 宮崎大学の論文
- 2010-09-30
著者
関連論文
- 曲面反射鏡の写像シミュレーションソフトの開発
- 曲面反射鏡の写像シミュレーションソフトの開発研究(県単共同研究)
- 基本周波数帯域におけるパワー減衰モデルを用いた音源数推定
- 改良型MSEPFによる画像からの物体追跡