 |
|
|
|
|
|
|
|
|
|
|
|
|
 |
|
|||||||||||
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
|
|
 |
 |
|
||||||||||
 |
|
|||||||||||
 |
|
|||||||||||
 |
 |
|
||||||||||
| Seminar Schedule (Seminar Organizer: Prof. Ziv Bar-Joseph) | |
Machine Learning Lunchtime Chats
|
Fall 2006 Seminar
Date: September 28, 2006 Time: 1:30 PM - 3:00 PM Location: 4623 Wean Hall Speaker: Fei-Fei Li Assistant Professor Title: Unsupervised Learning of Human Action Categories Using Spatial-Temporal Words Abstract: We present a novel unsupervised learning method for human action categories. A video sequence is represented as a collection of spatial-temporal words by extracting space-time interest points. The algorithm automatically learns the probability distributions of the spatial- temporal words and intermediate topics corresponding to human action categories. This is achieved by using a probabilistic Latent Semantic Analysis (pLSA) model. Given a novel video sequence, the model can categorize and localize the human action(s)contained in the video. We test our algorithm on two challenging datasets: the KTH human action dataset and a recent dataset of figure skating actions. Our results are on par or slightly better than the best reported results. In addition, our algorithm can recognize and localize multiple actions in long and complex video sequences containing multiple motions. Speaker Bio: Fei-Fei Li received her B.A. in Physics from Princeton University, along with two certificates in Engineering Physics, and Computational and Applied Math. She then spent a year in Tibet researching on Tibetan traditional medicine. In 2005, she obtained her Ph.D. in Electrical Engineering from California Institute of Technology. Fei-Fei is currently an assistant professor in Electrical and Computer Engineering Department, University of Illinois Urbana-Champaign. She will become anassistant professor in Computer Science at Princeton University starting January 2007. Fei-Fei's research interests are two-fold. In computer vision, she is interested in the problem of high-level recognition and semantic understanding of images and videos. Her current work involves Bayesian modeling for objects, scene and human action categorization. In human vision, she is particularly interested in natural scene recognition and the underlying cognitive and neural mechanisms, through psychophysics experiments.