In Computer Vision applications, there often exist the need to extract features to facilitate the process of analysis and recognition. In our key feature approach, we first study a computational framework of modeling human action styles, which extracts key motion trajectories for analysis and recognition assuming a given representation (e.g., trajectories and three-mode PCA). By emphasizing those key trajectories, the framework is capable of improving the modeling efficiency and accuracy, while being adaptive to multiple action style modeling criteria. We then extend the key feature approach in considering a general problem of recognition and analysis without an explicit representation (e.g., using raw images). We want to automatically extract key features to form an optimal internal representation of the data (images), which is a fundamental problem in Computer Vision research. By emphasizing the key information for representation, our approach is capable of improving the efficiency and accuracy for analysis and recognition in a wide range of applications in Computer Vision.
|School:||The Ohio State University|
|Department:||Computer and Information Science|
|School Location:||United States -- Ohio|
|Source:||DAI-B 79/09(E), Dissertation Abstracts International|
|Keywords:||Biological motion, Dimensionality reduction, Feature extraction, Linear discriminant analysis, Modeling human action styles, Three-mode pca|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be