Similar face Query Face Similar Face We propose a new method to retrieve similarface images from large face databases. The proposed method extracts a set ofHaar-like features, and integrates these features with supervised manifold learning. Haar-like features are intensity-basedfeatures. The values ofvarious Haar-like features comprise our rectangle feature vector (RFV) to describe faces. Compared with several popular unsupervised dimension reduction methods, RFV is more effective in retrieving similar faces. To further improve the performance, we combine RFV and a supervised manifold learning method and obtain satisfactory retrieval results.

We develop a new approach for gender recognition. In this paper, our approach uses the rectangle feature vector (RFV) as a representation to identify humans' gender from their faces. The RFV is computationally fast and effective to encode intensity variations of local regions of human face. By only using few rectangle features learned by AdaBoost, we present a gender identifier. We then use nonlinear support vector machines for classification, and obtain more accurate identification results.

No abstract

No abstract

This paper describes a tracking system for multiple moving objects undergoing a planar motion in a scene observed by a still camera. The objects are tracked through a sequence of frames. The direction and velocity of each object are calculated between each pair of frames and are used to predict the position of the object in the next frame. The calibration procedure and the problems due to the presence of noise and shadows in the images are also addressed and the adopted strategies are detailed. Our system has successfully been tested on a fish tracking application and is currently used to study the behavior of the fish in response to changes in environmental conditions.