Abstract. We present a novel technique for motion-based recognition of individual gaits in monocular sequences. Recent work has suggested that the image self-similarity plot of a moving person/object is a projection of its planar dynamics. Hence we expect that these plots encode much information about gait motion patterns, and that they can serve as good discriminants between gaits of different people. We propose a method for gait recognition that uses similarity plots the same way that face images are used in eigenface-based face recognition techniques. Specifically, we first apply Principal Component Analysis (PCA) to a set of training similarity plots, mapping them to a lower dimensional space that contains less unwanted variation and offers better separability of the data. Recognition of a new gait is then done via standard pattern classification of its corresponding similarity plot within this simpler space. We use the k-nearest neighbor rule and the Euclidian distance. We test this method on a data set of 40 sequences of six different walking subjects, at 30 FPS each. We use the leave-one-out crossvalidation technique to obtain an unbiased estimate of the recognition rate of 93%.

In this paper we present a real time pedestrian detection system that works on low qualio infrared videos. We intmduce pmbabilistic templates to capture the variations in human shape, especially for the case where contrast is low and body pans are missing. We present experimental results on infrared videos taken from a moving vehicle in various urban street scenarios to demonstrate the feasibility of the appmach.

No abstract

In this paper we present an ezample-based approach to learn a given class of wmplex shapes, and recognize instances ofthat shape with outliers. The system consists of a two-layer custom-designed neural network. We apply this appmch to the rewgnition ofpedestrians carrying objects from a single camem. The system is able to capture and model an ample mnge ofpedestrian shapes at varying poses and camem orientations, and achieves a 90% w m t recognition mte.

Many interactive multimedia applications require the ability to track the 3-D motion of participants in a room. Particle ¿lters are attractive for this since they do not require solution of the inverse problem of obtaining the state from measurements, and since the tracking can be easily extended to integrate multimodal measurements. We extend our previous work on smart videoconferencing to include a multimodal tracker of the session participants using multiple cameras and microphone arrays. We verify the correctness and robustness of the multimodal tracker using synthetic and real data. We also present practical details of how such a system can be implemented using off-the-shelf hardware and computers.