People
Projects
Software
Datasets
        
 
Dr. David Doermann
Faculty
Currently on Assignment to DARPA

Senior Research Scientist - Director of LAMP (1993 - 2014)
Ph D. 1993, Computer Science, University of Maryland

3451 AV Williams Building
College Park, MD 20742

Ph: 301-405-1767
Fax: 443-638-0236
Email: doermann@umiacs.umd.edu
Research Interests
Analysis and processing of document page images and analysis, indexing and retrieval of video information sources.
Projects
UMD Signature Matching/Verification Datasets
DIQA: Document Image Quality Assesment Datasets
Document Structure Learning Dataset
No-Reference Image Quality Assessment
Bengali Information Retrieval
Information Retrieval from Indic Script OCRed Text
DATASET: Rule-line removal in Handwritten Arabic Document Images
DATASET: Handwritten Arabic Textline Segmentation and Proximity Datasets
DATASET: LTP and Touching Text Segmentation
DATASET: Layer Separation - Arabic/English
MADCAT: Multilingual Automatic Document Classification Analysis and Translation
Page Segmentation
Recognition of People in Low Quality Video
Media Adaptation
Document Image Enhancement
Video Duplicate Detecton
Zone classification
Language and Script Identification
Signature Logo and Stamp Detection and Recognition
Vision on Mobile Devices
Compressed Domain Document Image analysis
Machine Printed Text and Handwriting Identification in Noisy Document Images
Rule Line Detection and Removal
Document Image Retreival
Document Image Classification
Signature Verification and Forgery Detection
DOCLIB - A C++ Library for Document Processing
GEDI: Groundtruthing Environment for Document Images (Software)
Functional Document Description
Duplicate Document Image Detection
Extraction of Text from Video
Enhancement of Text in Video
Mosaicing of MPEG Videos
Mosaicing from Camera Phones Images
ViPER: Video Performance Evaluation Resource
Video Segmentation
Video Classification
DATASET: Tobacco-800 Signatures and Logos
Map Analysis and Recognition Interface Application
Multi Classifier Combination
Camera Based Document Image Analysis
Hindi OCR
BRIDGE: Processing Bilingual Dictionaries
Video Content Visualization
Text Detection and Tracking in Digital Videos
Recent Publications

Sravanthi Bondugula, Varun Manjunatha, Larry S. Davis and David Doermann. "SHOE: Sibling Hashing with Output Embeddings." ACM Multimedia, 2015. (ACCEPTED). (BibTex) (Abstract) (PDF)

Rajiv Jain and David Doermann. "Localized Document Image Change Detection." International Conference on Document Analysis and Recognition, 2015. (SUBMITTED). (BibTex) (Abstract)

Le Kang, Peng Ye, Li Yi and David Doermann. "Simultaneous Estimation of Image Quality and Distortion via Multi-task Convolutional Neural Networks." The International Conference on Image Processing (ICIP 2015), September 2015. (BibTex) (Abstract)

Pawan Singh, Mita Nasipuri, Ram Sarkar and David Doermann. "Word-level Script Identification for Handwritten Indic scripts." International Conference on Document Analysis and Recognition, 2015. (SUBMITTED). (BibTex) (Abstract)

Hongxing Gao, Marc¸al Rusi˜nol, Dimosthenis Karatzas, Josep Llad´os, Rajiv Jain and David Doermann. "Novel Line Verification for Multiple Instance Focused Retrieval in Document Collections." International Conference on Document Analysis and Recognition, 2015. (SUBMITTED). (BibTex) (Abstract)

(Search for All Publications)


Last Updated: Saturday 17 May, 2014



home | language group | media group | sponsors & partners | publications | seminars | contact us | staff only
© Copyright 2001, Language and Media Processing Laboratory, University of Maryland, All rights reserved.