Suggested Term Projects CSE 666, Fall 2014 Guidelines • The described projects are suggestions; if you have desire, skills or idea to explore alternative topics, you are free to do so. • Finalize the project selection by October 16; have a 1-2 slide (2-3 minutes) presentation describing the project on that day. • Project reporting: - Final goal: estimate performance of proposed method or present a working demo of the system - Presentation: 20 minute on one of four days of classes(Nov. 20, 25, Dec. 2, 4). Presentation should include some topic background (previous works), description of the approach, preliminary results. - Final report : Dec. 16 ? – research style paper with results (background, previous work, detailed approach description, results, bibliography). • Working in groups: you can do so for large projects, but the work should be clearly split and separate reports from each person will have to be submitted. • Using existing projects, libraries, etc. You have to reference all such uses; the project should contain significant amount of your original work. Fingerprint Deformations • Skin Deformation Models for Fingerprint Matching -Implement previous methods for correlation matching J. Li, S. Tulyakov, and V. Govindaraju, "Verifying Fingerprint Match by Local Correlation Methods," in First IEEE International Conference on Biometrics: Theory, Applications, and Systems, 2007. BTAS 2007. , 2007, pp. 1-5. J. Li, S. Tulyakov, Z. Zhang, and V. Govindraju, "Fingerprint Matching Using Correlation and Thin-Plate Spline Deformation Model," in IEEE Second International Conference on Biometrics: Theory, Applications and Systems (BTAS 08), 2008. - Incorporate deformation cost into matching scores - Propose and implement other methods of skin deformation, e.g. R. Cappelli, D. Maio, and D. Maltoni. Modelling plastic distortion in Fingerprint images. In Proc. Int'l Conf. Advances in Pattern Recognition, 2001. Dongjin Kwon; Il Dong Yun; Sang-Uk Lee, "A Robust Warping Method for Fingerprint Matching," Computer Vision and Pattern Recognition, 2007. CVPR '07. IEEE Conference on , vol., no., pp.1,6, 17-22 June 2007 Fingerprint Indexing •Modify existing fingerprint indexing algorithm to use in-memory structures; test performance Mansukhani, P., Tulyakov, S., Govindaraju, V.: A Framework for Efficient Fingerprint Identification Using a Minutiae Tree. Systems Journal, IEEE 4 (2010) 126-137 • Implement alternative indexing method ; compare performances B. Bhanu and X. Tan, “Fingerprint indexing based on novel features of minutiae triplets,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 25, no. 5, pp. 616–622, May 2003. Cappelli, R.; Ferrara, M.; Maltoni, D., "Fingerprint Indexing Based on Minutia CylinderCode," Pattern Analysis and Machine Intelligence, IEEE Transactions on , vol.33, no.5, pp.1051,1057, May 2011 • Other indexing ideas? Fingerprint Image Binarization • Implement MRF based approach to fingerprint binarization: Huaigu Cao; Govindaraju, V., "Preprocessing of Low-Quality Handwritten Documents Using Markov Random Fields," Pattern Analysis and Machine Intelligence, IEEE Transactions on , vol.31, no.7, pp.1184,1194, July 2009 Kai Cao; Eryun Liu; Jain, AK., "Segmentation and Enhancement of Latent Fingerprints: A Coarse to Fine Ridge Structure Dictionary," Pattern Analysis and Machine Intelligence, IEEE Transactions on , vol.36, no.9, pp.1847,1859, Sept. 2014 Fingerprint Hashing • Implement some fingerprint hashing algorithms and compare their performance with existing methods Farooq, F., Bolle, R.M., Jea, T.-Y., Ratha, N.A.-R., N.: Anonymous and Revocable Fingerprint Recognition. In: Bolle, R.M. (ed.): Computer Vision and Pattern Recognition, 2007. CVPR '07. IEEE Conference on (2007) 1-7 Ferrara, M.; Maltoni, D.; Cappelli, R., "Noninvertible Minutia Cylinder-Code Representation," Information Forensics and Security, IEEE Transactions on , vol.7, no.6, pp.1727,1737, Dec. 2012 Face Tracking • Improve face tracking accuracy by performing direction dependent recognition of face parts - Need to re-train algorithms (SVM) performing recognition - Data for training can be collected automatically based on currently recognized positions of recognized patches. Some data verification might be needed. • Improve face tracking accuracy by utilizing non-rectangular region recognition •Improve face tracking speed by utilizing alternative features (e.g. LBP instead of HOG), feature reduction methods, SVM replacement? •Improving face tracking algorithm by adding user-specific matching Dalal, N.; Triggs, B.; , "Histograms of oriented gradients for human detection," Computer Vision and Pattern Recognition, 2005. CVPR 2005. IEEE Computer Society Conference on , vol.1, no., pp.886-893 vol. 1, 25-25 June 2005 M. Isard and A. Blake, “ICONDENSATION- Unifying low-level and high-level tracking in a stochastic framework,” Proceedings of European Conference on Computer Vision, Vol. 1, pp. 893-908, 1998. Facial Expression Matching • Automated facial expression analysis : - Given positions of eyes, brows and mouth in videos - Detect head movements (e.g. speed in movements, direction, etc.) and investigate if they can be used for person identification -Detect facial expressions from the changes in features extracted near landmark points - Explore if sequences of changes can be used for person identification S. Tulyakov, T. Slowe, Z. Zhang, and V. Govindaraju, "Facial Expression Biometrics Using Tracker Displacement Features," in Computer Vision and Pattern Recognition, 2007. CVPR '07. IEEE Conference on, 2007, pp. 1-5. A. L. Kashyap, S. Tulyakov, and V. Govindaraju, "Facial Behavior as a Soft Biometric," in International Conference on Biometrics (ICB 2012), 2012. Multi-frame face matcher score fusion • Match subsequent frames in the video to a single enrolled face image • Need to effectively fuse these scores - Use difference between faces in video frames - Goal: automatically learnable fusion method S. P. Satheesan, S. Tulyakov, and V. Govindaraju, "A feature information based approach for enhancing score-level fusion in multi-sample biometric systems," in Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG), 2013 Fourth National Conference on, 2013, pp. 1-4. X. Cheng, S. Tulyakov, and V. Govindaraju, "Utilization of matching score vector similarity measures in biometric systems," in Computer Vision and Pattern Recognition Workshops (CVPRW), 2012 IEEE Computer Society Conference on, 2012, pp. 111-116.