Robustness and tracking speed are two important indices for evaluating the performance of real-time 3-D tracking. We propose a new approach to fuse sensing data of the most current observation into a 3-D visual tracker with particle techniques. With the proposed data fusion method, the importance density function in the particle filter can be designed to represent posterior states by particle crowds in a better way. This makes the tracking system more robust to noise and outliers. On the other hand, because particle interpretation is performed in a much more efficient fashion, the number of particles used in tracking is greatly reduced, which improves the real-time performance of the system. Simulation and experimental results verified the effectiveness of the proposed method.