UT CS 395T - Efficient Learning of Relational Object Class Models (8 pages)

Previewing pages 1, 2, 3 of 8 page document View the full content.
View Full Document

Efficient Learning of Relational Object Class Models



Previewing pages 1, 2, 3 of actual document.

View the full content.
View Full Document
View Full Document

Efficient Learning of Relational Object Class Models

90 views

Lecture Notes


Pages:
8
School:
University of Texas at Austin
Course:
Cs 395t - Multicore Operating Systems Implementation
Multicore Operating Systems Implementation Documents

Unformatted text preview:

In Proc 10th IEEE International Conference of Computer Vision beijing October 2005 Efficient Learning of Relational Object Class Models Aharon Bar Hillel Tomer Hertz Daphna Weinshall School of Computer Science and Engineering and the Center for Neural Computation Hebrew university of Jerusalem Israel 91904 aharonbh tomboy daphna cs huji ac il Abstract represent both parts appearance and invariant relations of location and scale between the parts Part based models are somewhat resistant to various sources of variability such as within class variance partial occlusion and articulation and they may be convenient for indexing in a more complex system We present an efficient method for learning part based object class models The models include location and scale relations between parts as well as part appearance Models are learnt from raw object and background images represented as an unordered set of features extracted using an interest point detector The object class is generatively modeled using a simple Bayesian network with a central hidden node containing location and scale information and nodes describing object parts The model s parameters however are optimized to reduce a loss function which reflects training error as in discriminative methods Specifically the optimization is done using a boosting like technique with complexity linear in the number of parts and the number of features per image This efficiency allows our method to learn relational models with many parts and features and leads to improved results when compared with other methods Extensive experimental results are described using some common bench mark datasets and three sets of newly collected data showing the relative advantage of our method Part based approaches to object class recognition can be crudely divided into two types 1 generativemodel based methods e g 11 and 2 discriminativemodel free methods e g 2 In the GenerativeFigure 1 Dog immodel based approach a age with our learnt model drawn



View Full Document

Access the best Study Guides, Lecture Notes and Practice Exams

Loading Unlocking...
Login

Join to view Efficient Learning of Relational Object Class Models and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Efficient Learning of Relational Object Class Models and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?