Gait Recognition

The goal of this research is to recognize (identify) people using relatively low resolution video sequences of people walking (e.g., in a hallway, across a parking lot). We use two approaches: The first is a parametric approach, in which we estimate biometrics such as stride length, cadence, and height. This approach can be view invariant, but requires a calibrated camera system. In the second approach, we extract a non-parametric projection of a person’s phase space, and use PCA analysis for recognition. This approach is view dependent, but does not require a calibrated camera system. These two gait recognition techniques can be combined to provide improved accuracy.

Related Publications

Chiraz BenAbdelkader, Ross Cutler and Larry Davis. “Gait Recognition Using Image Self-Similarity.” EURASIP Journal on Applied Signal Processing, 2004, 4 (2004) 572-585. PDF.

Chiraz BenAbdelkader, Ross Cutler, and Larry Davis. “Person Identification using Automatic Height and Stride Estimation”, IEEE International Conference on Pattern Recognition, 2002. PDF.

Chiraz BenAbdelkader, Ross Cutler, and Larry Davis, “View-invariant estimation of height and stride for gait recognition”, Workshop on Biometric Authentication (BIOMET) 2002, in association with ECCV 2002. Compressed Postscript.

Chiraz BenAbdelkader, Ross Cutler, and Larry Davis, “Stride and Cadence as a Biometric in Automatic Person Identification and Verification”, 5th International Conference on Automatic Face and Gesture Recognition, 2002. PDF.

Chiraz BenAbdelkader, Ross Cutler and Larry Davis, “Motion-based Recognition of People in Eigengait Space”, 5th International Conference on Automatic Face and Gesture Recognition, 2002. PDF.

C. BenAbdelkader, Ross Cutler, Harsh Nanda, and L. S. Davis, “EigenGait: Motion-based Recognition of People using Image Self-similarity”, Proc. Intl Conf. on Audio and Video-based Person Authentication (AVBPA), 2001. PDF.