Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE credits
TLD is an award-winning, real-time algorithm for long-term tracking of unknown objects in
video streams. The object of interest is defined by a bounding box in a single frame. TLD
tracks the object, Learns its appearance and Detects it whenever it appears
in the video. The result is a real-time tracking that typically improves over time. Long-term
tracking of arbitrary objects is a the core problem in many computer vision applications:
surveillance, object auto-focus, SLAM, games, HCI, video annotation etc.
The following work is done:
1. The OpenTLD algorithm is extended for Multi-Object Tracking.
2. The OpenTLD algorithm is evaluated (i.e. problems with the algorithm, usability of
the algorithm), specially from Eye tracking perspective.
3. The OpenTLD algorithm is compared with other tracking algorithm e.g. Mean shift
in OpenCV in terms of tracking performance.
4. Some enhancements to the OpenTLD algorithm are made.