Tutorial: The Open Set Recognition Problem in Information Forensics and Security

2015 IEEE Intl. Workshop on Information Forensics and Security (WIFS), Rome, Italy

Introduction | Lecturers | Presentation Slides | Source-codes | Additional Reading

Tutorial Fast-track Introduction

Coinciding with the rise of large-scale statistical learning within the information forensics and security community there has been a dramatic improvement in automated methods for digital image forensics, forensic linguistics, network intrusion detection, and human biometrics, among many other applications. Despite this progress, a tremendous gap exists between the performance of automated methods in the laboratory and the performance of those same methods in the field. A major contributing factor to this is the way in which machine learning algorithms are typically evaluated: without the expectation that a class unknown to the algorithm at training time will be experienced during operational deployment. The purpose of this tutorial is to introduce the WIFS audience to the open set recognition problem. A number of different topics will be explored, including supervised machine learning, probabilistic models, kernel machines, the statistical extreme value theory, and original case studies related to the topics of interest at WIFS. The tutorial is composed of three parts, each lasting approximately one hour. Part #1: An introduction to the open set recognition problem? Part #2: Algorithms that minimize the risk of the unknown Part #3: Case studies related to images, text and network traffic

Up

Lecturers

Walter J. Scheirer (University of Notre Dame, IN, US) Walter J. Scheirer, Ph.D. is an Assistant Professor in the Department of Computer Science and Engineering at the University of Notre Dame. Prior to Notre Dame, he was a Postdoctoral Fellow in the Center for Brain Science at Harvard University. He received his Ph.D. from the University of Colorado and his M.S. and B.A. degrees from Lehigh University. Dr. Scheirer has extensive experience in the areas of security, computer vision and human biometrics, with an emphasis on advanced learning techniques. His overarching research interest is the fundamental problem of recognition, including the representations and algorithms supporting solutions to it.

Anderson Rocha (University of Campinas (Unicamp), SP, Brazil) Anderson Rocha received his B.Sc degree from Federal University of Lavras, Brazil in 2003. He received his M.S. and Ph.D. from University of Campinas, Brazil in 2006 and 2009, respectively. Currently, he is an Associate Professor in the Institute of Computing, Unicamp, Brazil. His main interests include digital image and video forensics, pattern analysis, machine learning, and reasoning for complex data. He is an elected member of the IEEE IFS-TC and was co-general chair of WIFS 2011. In 2011, he was named a Microsoft Research Faculty Fellow. Finally, he is an associate editor for IEEE T-IFS.

Up

Presentation Slides

Up

Source-codes

The source-code for W-SVM, PI-SVM, and 1-vs-Set is available directly from GitHub.

(HTML)

DBC Decision Boundary Carving -- It requires LibSVM.

31 KB

(ZIP)

Up

Additional Reading

F. Costa, E. Silva, M. Eckmann, W.J. Scheirer, and A. Rocha. Open Set Source Camera Attribution and Device Linking, Pattern Recognition Letters, 2014. W.J. Scheirer, A. Rocha, A. Sapkota, and T. Boult. Towards Open Set Recognition, IEEE T-PAMI, 35(7) July 2013. M.J. Wilber, W.J. Scheirer, P. Leitner, B. Heflin, J. Zott, D. Reinke, D. Delaney, T.E. Boult. Animal Recognition in the Mojave Desert: Vision Tools for Field Biologists, IEEE WACV, 2013. B. Heflin, W.J. Scheirer, and T.E. Boult. Detecting and Classifying Scars, Marks, and Tattoos Found in the Wild, IEEE BTAS, 2012. W.J. Scheirer, A. Rocha, R. Micheals, and T.E. Boult. Meta-Recognition: The Theory and Practice of Recognition Score Analysis, IEEE T-PAMI, 33(8), 2011. A. Rattani, W.J. Scheirer, and A. Ross. Open Set Fingerprint Spoof Detection Across Novel Fabrication Materials, IEEE T-IFS, 10(11) Nov. 2015. W.J. Scheirer, L.P. Jain, and T.E. Boult. Probability Models for Open Set Recognition, IEEE T-PAMI, 36(11), Nov. 2014. L.P. Jain, W.J. Scheirer, and T.E. Boult. Multi-class Open Set Recognition Using Probability of Inclusion, ECCV, Sept. 2014.

Up