The condensation tracking algorithm uses a prior transition probability as the proposal distribution, which does not make full use of the current observation. In order to overcome this shortcoming, a new face tracking algorithm based on particle filter with mean shift importance sampling is proposed. First, the coarse location of the face target is attained by the efficient mean shift tracker, and then the result is used to construct the proposal distribution for particle propagation. Because the particles obtained with this method can cluster around the true state region, particle efficiency is improved greatly. The experimental results show that the performance of the proposed algorithm is better than that of the standard condensation tracking algorithm.
An object learning and recognition system is implemented for humanoid robots to discover and memorize objects only by simple interactions with non-expert users. When the object is presented, the system makes use of the motion information over consecutive frames to extract object features and implements machine learning based on the bag of visual words approach. Instead of using a local feature descriptor only, the proposed system uses the co-occurring local features in order to increase feature discriminative power for both object model learning and inference stages. For different objects with different textures, a hybrid sampling strategy is considered. This hybrid approach minimizes the consumption of computation resources and helps achieving good performances demonstrated on a set of a dozen different daily objects.
