Home Page of Jayant Kumar

Jayant Kumar 3348, A. V. Williams Bldg. Department of Computer Science, University of Maryland, College Park, MD Email: my_first_name AT umiacs DOT umd DOT edu Resume: jayant_kumar_cv Profiles: Google scholar, DBLP, LinkedIn, GitHub
I am a computer vision and machine learning researcher at PARC, Xerox Research (NY). Previously, I completed my PhD under the supervision of Dr. David Doermann and Prof. Larry Davis at Univ. of Maryland, College Park. My research interests lie in the areas of mobile and wearable computer vision and machine learning applied to image/video understanding. Currently I am involved in an Explore project dealing with mobile sensing and analytics of human behavior. I am a primary collaborator on a UAC project on Egocentric Vision and sensing with Prof. Jim Rehg. I am a secondary collaborator in other UAC project on computational photography for document images with UMD.
Education: Ph.D., Computer Science, Univ. of Maryland College Park, 2008-2013 (Thesis: Efficient Machine Learning methods for Document Image Analysis) M.S., Computer Science, Univ. of Maryland College Park, 2008-2010 B.E., Computer Science & Engineering, R. V. College of Engineering, Bangalore, 2002-2006
Work Experience: Research Scientist, PARC, A Xerox company, Webster, NY, Oct. 2013 - Present Research Assistant, Language and Media Processing Lab. ,UMIACS, College Park, August 2008 - December 2013 Intern , Xerox Innovation group , Xerox Research Center Webster , Webster, May 2012 - August 2012 Intern, Multimedia systems, Fuji-Xerox Palo Alto Lab , Palo Alto, May 2011 - August 2011 Intern, Speech, Language and Multimedia group , Raytheon BBN technologies , Boston, June 2010 - Aug. 2010 Research Assistant, MILE Lab., Indian Institute of Science, Bangalore, Feb. 2007 - June 2008 Development Engineer, Aditi Technologies, Bangalore, July 2006 - Feb. 2007 Intern, NonStop Enterprise Division, Hewlett-Packard, Bangalore, Jan. 2006 - May 2006
Publications Mobile and wearable vision, Image quality, Sharpness estimation On-the-Fly Hand Detection Training with Application in Egocentric Action Recognition, IEEE CS Workshop on Observing and understanding hands in action (HANDS with CVPR 15) , 2015. [J. Kumar, Q. Li, S. Kyal, E. Bernal and R. Bala] Flash/No-Flash Fusion for Mobile Document Image Binarization, IEEE Intl. Conf. on Image Processing (ICIP) , 2014. [J. Kumar, M. Maltz and R. Bala] Beyond Human Opinion Scores: Blind Image Quality Assessment based on Synthetic Scores, IEEE Conf. on Computer Vision and Pattern Recognition (CVPR) , 2014. [P. Ye, J. Kumar and D. Doermann] Mobile Video Capture of Multi-page Documents, Intl. Workshop on Mobile Vision (workshop of CVPR), 2013. (Oral) [J. Kumar, R. Bala, H. Ding and P. Emmett] Real-time No-Reference Image Quality Assessment based on Filter Learning, IEEE Conf. on Computer Vision and Pattern Recognition (CVPR) , 2013. [P. Ye, J. Kumar, L. Kang and D. Doermann] Unsupervised Feature Learning Framework for No-reference Image Quality Assessment, Intl. Conf. on Computer Vision and Pattern Recognition (CVPR), 2012. (MATLAB Code) [P. Ye, J. Kumar, L. Kang and D. Doermann] Sharpness Estimation for Document and Scene Images, Intl. Conf. on Pattern Recognition (ICPR) , 2012. (Oral) [J. Kumar, F. Chen and D. Doermann] SmartDCap: Semi-Automatic Capture of Higher Quality Document Images from a Smartphone, Intl. Conf. on Intelligent User Interfaces (IUI) , 2013. (Oral) [F. Chen, S. Carter, L. Denoue and J. Kumar] A Dataset for Quality Assessment of Camera Captured Document Images, Camera Based Document Analysis and Recognition (workshop of ICDAR), 2013. [J. Kumar, P. Ye and D. Doermann] Structural similarity, Retreival, Classification Convolutional Neural Networks for Document Image Classification, Intl. Conf. on Pattern Recognition (ICPR) 2014 [L. Kang, J. Kumar, P. Ye, Y. Li and D. Doermann] Unsupervised Classification of Structurally Similar Document Images, Intl. Conf. on Document Analysis and Recognition (ICDAR) , 2013. (poster) [J. Kumar and D. Doermann] Structural Similarity for Document Image Classification and Retrieval, Pattern Recognition Letters (PRL) ,Special Issue, 2013 [J. Kumar, P. Ye and D. Doermann] Learning Document Structure for Retrieval and Classification, Intl. Conf. on Pattern Recognition (ICPR) , 2012.(Oral) [Best student paper award] [J. Kumar, P. Ye and D. Doermann] Document Image Classification and Labeling using Multiple Instance Learning. Intl. Conf. on Document Analysis and Recognition (ICDAR), 2011. (Oral) [J. Kumar, J. Pillai and D. Doermann] Text-line Segmentation, Text-zone classification, Text Binarization Learning Text-line Segmentation using Codebooks and Graph Partitioning, Intl. Conf. on Frontiers in Handwriting Recognition (ICFHR) , 2012. (Oral) [L. Kang, J. Kumar, P. Ye and D. Doermann] Segmentation of Handwritten Textlines in Presence of Touching Components, Intl. Conf. on Document Analysis and Recognition (ICDAR), 2011. [J. Kumar, L. Kang, D. Doermann and W. Abd-Almageed] Handwritten Arabic Text Line Segmentation using Affinity Propagation. Document Analysis Systems (DAS) , 2010. (Oral) [Among top 10 papers] [J. Kumar, W. Abd-Almageed, L. Kang and D. Doermann] Shape Codebook based Handwritten and Machine Printed Text Zone Extraction. Document Recognition and Retrieval(DRR), 2011. (Oral) [J. Kumar, R. Prasad, H. Cao, W. Abd-Almageed, D. Doermann and P. Natarajan] Font and Background Color Independent Text Binarization . Camera Based Document Analysis and Recognition (workshop of ICDAR), 2007. (Oral) [T. Kasar, J. Kumar and A. G. Ramakrishnan] Pre-processing, Rule-line detection and removal Fast Rule-line Removal using Integral Images and Support Vector Machines. Intl. Conf. on Document Analysis and Recognition (ICDAR), 2011. (Oral) [J. Kumar and D. Doermann] Page Rule-Line Removal using Linear Subspaces in Monochromatic Handwritten Arabic Documents. Intl. Conf. on Document Analysis and Recognition (ICDAR), 2009. [W. Abd-Almageed, J. Kumar and D. Doermann] Line removal and Restoration of Handwritten strokes. Intl. Conf. on Computational Intelligence and Multimedia Applications (ICCIMA), Sivakasi, 2007.(Oral) [K.R. Arvind, J. Kumar and A. G. Ramakrishnan]
Patents: Sharpness Estimation in Document and Scene Images (US patent, Pub. date: 2013/5/16, Patent no.: 20130121610) Smart document capture based on likely scanned-image quality (US patent filed Aug. 2012 with FXPAL) Video Capture of Multi-faceted documents (US patent filed Dec. 2012 with Xerox Corp.)
Graduate Courses (by area): Computer Vision: Image Understanding, Image Segmentation, Object Recognition AI and Learning: Machine Learning, Neural Computation, Computational Linguistics II, Link Mining (Entity Resolution) Algorithms: Design and Analysis of Computer Algorithms, Computational Geometry, Scientific Computing I Software Engineering: Fundamentals of Software Testing
Last updated: Aug. 29, 2013