latest news

01.05.2008

Files for In-Class Exercise on 01.04.2008 are now posted.

read more ...

12.05.2007

Doctoral standing is not required to register for this course.

read more ...

other information

Instructor: Shishir Shah

Office: PGH 564

Office Hours: M 12:00-1:00pm

Class Location: PGH 232

Class Hours: M 1:00-4:00pm

Phone: 713-743-3360

Email: shah@cs.uh.edu

TA: Apurva Gala

TA Hours: TBA

TA Office: PGH 550E

Email: avbedagk@mail.uh.edu

Homework Assignments

ASSIGNMENTPROBLEMNOTESDUE DATEFILES
HW 1Probabilistic Image Segmentation OverviewMarch 4, 2008, 11:59pmHW1 Files
HW 2Background Subtraction-April 25, 2008, 11:59pmHW2 Files, Uncompressed Videos, Images from Mall Sequence, Images from Outdoor Sequence.

Term Projects

The purpose of the term project is to expose the students to the realities of designing, implementing and evaluating pattern recognition systemss for real-world problems. The objective is to consider the various stages of any pattern recognition system, namely feature extraction and selection, model learning and estimation, and classifier generation and evaluation. Various algorithms should be considered and evaluated quantitatively to identify the most suited for the problem at hand, with the eventual goal of maximing classification accuracy while minimizing error for the selected datasets.

You can use your own data from your research or select datasets from various public resources listed on the resources page. You can also contact the instructor if specific image and video datasets. You are free to use any programming language. You can write the codes yourself or use any code that is available in the public domain. In case you use public domain source code, please cite the appropriate reference and copyright. You are required to know the details of the algorithm that the public domain code implements. In all case, prior approval of the project scope should be obtained before starting your project.

You are required to work as a group of 3-4 members. Submissions related to the project will include a project proposal, an interim progress report, a final report written in a conference paper format, and make a oral presentation during the finals week. Tentative schedule of the project is as follows:

DATEMILESTONEDESCRIPTION
March 10Project ProposalSubmit a 1-2 page proposal that describes the problem you would like to tackle, objective of the study, data that you plan to utilize, and evaluation strategies that you plan to use. Also provide a short list of related references.
April 7Progress ReportSubmit a report that describes your progress with the project and your plans for the rest of the semester.
April 28Oral PresentationMake a 15 minute presentation of your work to the class.
May 5Final ReportSubmit a readable and well-organized report that provides proper motivation for the task, proper citation and discussion of related literature, proper explanation of the details of the approach and implementation strategies, proper performance evaluation, and detailed discussion of the results. Highlight your contributions and conclusions. A well-documented software should be submitted with your report. Each team member should also provide a written description of her/his own contributions to the project.

Final Report Guidelines:

The final report should follow the IEEE two-column format. The report should be no more than 12 pages in length. Please use the IEEE's LaTex template or the Word template.

Possible Project Topics:

  1. Image Segmentation
  2. Object Recognition in Natural Scenes
  3. Make and Model Recognition of a Car
  4. Reading License Plate from Video Imagery
  5. Comparison of Unsupervised Learning Algorithms
  6. Comparison of Classification Algorithms on Handwritten Data
  7. Classification of Spatial Patterns in Time
  8. Object Classification using Boosted Classifiers
  9. Content Modeling using Probabilistic Learning

Projects and Teams:

  1. Image Segmentation using Graphical Models - Benjamin Soibam, Ning Situ, Xuqing Wu, and Zhengnan Wang
  2. Automatic Image Labeling for Content Based Image Retrieval - Akash Bhatt, Chaitanya Bagaria, Nripun Sredar, and Raja Yalamanchili
  3. Blood Classification in Intravascular Ultrasound Using Boosted Support Vector Machine - Eleni Sgouritsa, Feki Saber, and Hakan Haberdar
  4. License Plate Recognition - Keith Lancaster, Tarun Wadhawan, Yan Zhou, and Yuichi Fujiki
  5. Music Classification System - Krushita Shah, Ashish Kapadia, Gezla Gafoor, and Santosh Sagar
  6. Automatic Car Detection - Weilong Yang, Wei Song, Zhijang Qiao, and Tianhong Fang