Pfinder: Real-Time Tracking of the Human Body
IEEE Transactions on Pattern Analysis and Machine Intelligence
W4: Real-Time Surveillance of People and Their Activities
IEEE Transactions on Pattern Analysis and Machine Intelligence
Generalized Parallel-Perspective Stereo Mosaics from Airborne Video
IEEE Transactions on Pattern Analysis and Machine Intelligence
Optical flow-based tracking of deformable objects using a non-prior training active feature model
PCM'04 Proceedings of the 5th Pacific Rim conference on Advances in Multimedia Information Processing - Volume Part III
Panoramic appearance-based recognition of video contents using matching graphs
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Hi-index | 0.00 |
Active Shape Model (ASM) paradigm is a popular method for image segmentation where a priori information about the shape of the object of interest is available. The effectiveness of the method is contingent upon a correct correspondence between model points and the features extracted from the image. Extensive application of these models soon revealed one of their limitations when, for a given model point, no obvious salient point can be found in the image. The primary cause of such limitation is due to weak edges and presence of abrupt noise which is the case with low light surveillance video images. In this paper we propose a fusion-based panoramic tracking algorithm of in low light images using multiple sensors. The proposed algorithm uses an IR and CCD sensor for image capture. The proposed tracking system consists of three steps: (i) pyramid based fusion algorithm, (ii) reconstruction of panoramic image, and (iii) active shape model (ASM)-based tracking algorithm. The experimental results show that the proposed tracking system can robustly extract and track objects on panoramic images in real-time.