We present a multiple classifier system for model-free tracking. The tasks of detection (finding the object of interest), recognition (distinguishing similar objects in a scene), and tracking (retrieving the object to be tracked) are split into separate classifiers in the spirit of simplifying each classification task. The supervised and semi-supervised classifiers are carefully trained on-line in order to increase adaptivity while limiting accumulation of errors, i.e. drifting. In the experiments, we demonstrate real-time tracking on several challenging sequences, including multi-object tracking of faces, humans, and other objects. We outperform other on-line tracking methods especially in case of occlusions and presence of similar objects.

We propose a novel approach to increase the robustness of object detection algorithms in surveillance scenarios. The cascaded confidence filter successively incorporates constraints on the size of the objects, on the preponderance of the background and on the smoothness of trajectories. In fact, the continuous detection confidence scores are analyzed locally to adapt the generic detector to the specific scene. The approach does not learn specific object models, reason about complete trajectories or scene structure, nor use multiple cameras. Therefore, it can serve as preprocessing step to robustify many tracking-by-detection algorithms. Our real-world experiments show significant improvements, especially in the case of partial occlusions, changing backgrounds, and similar distractors.

A fundamental problem of object tracking is to adapt to unseen views of the object while not getting distracted by other objects. We introduce Dynamic Objectness in a discriminative tracking framework to sporadically rediscover the tracked object based on motion. In doing so, drifting is eectively limited since tracking becomes more aware of objects as independently moving entities in the scene. The approach not only follows the object, but also the background to not easily adapt to other distracting objects. Finally, an appearance model of the object is incrementally built for an eventual re-detection after a partial or full occlusion. We evaluated it on several well-known tracking sequences and demonstrate results with superior accuracy, especially in dicult sequences with changing aspect ratios, varying scale, partial occlusion and non-rigid objects.

Digital impressions of teeth, obtained through intra-oral scanning, allow for more efficient and cost effective treatments of many dental indications. Current state-of-the-art intra-oral impression acquisition systems make use of a separate monitor to show the scanning progress, forcing the dentist to divert attention away from the scanner and the patient. In this paper, we present an augmented reality based solution to this problem. During the scanning process, an optical see-through head-mounted display is used to show an online overlay of the dynamic dental model onto the patient's teeth. The dentist can then fully focus on the patient and the scanner, while still being able to keep track of the current state of the model. This type of novel application, which fundamentally changes the humancomputer interaction of intra-oral scanning systems, requires a fast and accurate registration of a dynamically growing model onto a glossy, partially occluded surface at a very small scale. To meet this demand, we propose application tailored algorithms for indirect high accuracy online 3D teeth tracking and optical see-through head-mounted display calibration. Experimental results indicate that our system does have a potential to noticeably facilitate intraoral scanning in the future.

No abstract