Students should consult this page at least once a week for important information.
For next week, be sure to read the HMM background paper. I can't find it on-line, so come by my office to pick a copy sometime.
Here are the questions from this week:
1. How is CONDENSATION tracking able to handle temporary occlusions?
2. When considering composite gestures, the authors compute the probability of a parent gestures by summing the probabilities of its component gestures. What is the limitation of this method if we want to deal with more complicated composite gestures?
Location: Pratt 378, Wednesday 1-3
Instructor: Richard Zemel, email: zemel@cs
Office: Pratt 265D, 978-7497
Office Hours: Tuesday 1-3,
or make an appointment, or send an email message.
Further information for the course is available through these links: