Daniel Lopresti | Lehigh University (original) (raw)

Uploads

Papers by Daniel Lopresti

Research paper thumbnail of On-line handwriting recognition

Skip to Main Content. ...

Research paper thumbnail of A National Discovery Cloud: Preparing the US for Global Competitiveness in the New Era of 21st Century Digital Transformation

arXiv (Cornell University), Apr 14, 2021

Research paper thumbnail of CCC/Code 8.7: Applying AI in the Fight Against Modern Slavery

ArXiv, 2021

On any given day, tens of millions of people find themselves trapped in instances of modern slave... more On any given day, tens of millions of people find themselves trapped in instances of modern slavery. The terms"human trafficking,""trafficking in persons,"and"modern slavery"are sometimes used interchangeably to refer to both sex trafficking and forced labor. Human trafficking occurs when a trafficker compels someone to provide labor or services through the use of force, fraud, and/or coercion. The wide range of stakeholders in human trafficking presents major challenges. Direct stakeholders are law enforcement, NGOs and INGOs, businesses, local/planning government authorities, and survivors. Viewed from a very high level, all stakeholders share in a rich network of interactions that produce and consume enormous amounts of information. The problems of making efficient use of such information for the purposes of fighting trafficking while at the same time adhering to community standards of privacy and ethics are formidable. At the same time they help us,...

Research paper thumbnail of An Architecture for Programmable Systolic Arrays

Systolic algorithms are a systematic and powerful methodology for realizing the potential of very... more Systolic algorithms are a systematic and powerful methodology for realizing the potential of very large scale integration. After an initial wave of special-purpose systolic processors, research has turned to the implementation of programmable systolic machines. These machines are often too large and complicated to be able to support massive parallelism: many of the benefits to be reaped from systolic algorithms are lost. Few of the current architectures provide the simple systolic communication of the early special-purpose processors. The primary contribution of this research is the systolic shared register paradigm. Unlike previous research efforts, the architecture provides a natural means of systolic communication and uses simple processing elements; the architecture can be both easily programmed and easily formed into very large systolic arrays. The main features of the architecture are broadcast instructions, a regular topology, and register sets that are shared between neighbo...

Research paper thumbnail of Reproducible Research in Pattern Recognition, Second International Workshop, RRPR 2018

Preface This volume contains the articles from the second edition of the Workshop on Reproducible... more Preface This volume contains the articles from the second edition of the Workshop on Reproducible Research in Pattern Recognition that was held during August 20, 2018, in conjunction with ICPR 2018 in Beijing. It followed in the same spirit as the first edition with a special focus on digital geometry and mathematical morphology. It was intended as both a short participative course on reproducible research (RR) aspects leading to open discussions with the participants and also on how to actually perform RR. For this second edition, a new call for short papers was proposed to the ICPR authors. The main idea was to give authors of already accepted ICPR papers the possibility to highlight the reproducibility of their work with a companion paper. It was an opportunity to include implementation details, source code descriptions, parameter choice etc. This proceedings volume gathers 14 contributions covering the RR result track (three papers), invited RR contributions (five papers), and t...

Research paper thumbnail of An Episodic Learning Network for Text Detection on Human Bodies in Sports Images

IEEE Transactions on Circuits and Systems for Video Technology, 2021

Due to the proliferation of sports-related multimedia content on the WWW, effective visual search... more Due to the proliferation of sports-related multimedia content on the WWW, effective visual search and retrieval present interesting research challenges. These are caused by poor image quality, a wide range of possible camera points of view, pose variations on the part of athletes engaged in playing a sport, deformations of text appearing on sports person’s clothing and uniforms in motion, occlusions caused by other objects, etc. To address these challenges, this paper presents a new method for detecting text on human bodies in sports images. Unlike most existing methods, which attempt to exploit locations of a player’s torso, face, and skin, we propose an end-to-end episodic learning approach that employs inductive learning criteria for detecting clothing regions in an image, which are, in turn, then used for text detection. Our method integrates a Residual Network (ResNet) and Pyramidal Pooling Module (PPM) for generating a spatial attention map. The Progressive Scalable Expansion Algorithm (PSE) is adapted for text detection from these regions. Experimental results on our own dataset as well as several benchmarks (like RBNR and MMM which contain images of runners in marathons, and Re-ID which is a person re-identification dataset) demonstrate that the proposed method outperforms existing methods in terms of precision and F1-score. We also present results for sports images chosen from natural scene text detection datasets such as CTW1500 and MS-COCO to show the proposed method is effective and reliable across a range of inputs.

Research paper thumbnail of A New Context-Based Method for Restoring Occluded Text in Natural Scene Images

Document Analysis Systems, 2020

Research paper thumbnail of Proceedings of the 5th International Workshop on Document Analysis Systems V

Research paper thumbnail of Document Recognition V

Document Recognition V, Apr 1, 1998

Research paper thumbnail of Document Recognition and Retrieval VII

Document Recognition and Retrieval Vii, Dec 1, 1999

Research paper thumbnail of Ink Matching of Cursive Chinese Handwritten Annotations

Series in Machine Perception and Artificial Intelligence, 1999

Research paper thumbnail of Robust retrieval of noisy text

Proceedings of the Third Forum on Research and Technology Advances in Digital Libraries,

Research paper thumbnail of The Role of Document Image Analysis in Trustworthy Elections

Statistical Science and Interdisciplinary Research, 2014

Research paper thumbnail of Ink as a First-Class Datatype in Multimedia Databases

Artificial Intelligence, 1996

Research paper thumbnail of Resource-optimized delivery of web images to small-screen devices

Research paper thumbnail of <title>OCR for World Wide Web images</title>

Document Recognition IV, 1997

Research paper thumbnail of Supporting experimental research in computer vision

Research paper thumbnail of The Ninth IAPR International Workshop on Document Analysis Systems, DAS 2010, June 9-11, 2010, Boston, Massachusetts, USA

Research paper thumbnail of Discounts for dynamic programming with applications in VLSI processor arrays

Description/Abstract This dissertation introduces a method for transforming certain dynamic progr... more Description/Abstract This dissertation introduces a method for transforming certain dynamic programming problems into ones that require less space and time to solve under the logarithmic cost criterion, an appropriate complexity measure for flexible word-length ...

Research paper thumbnail of Delta transformations to simplify VLSI processor arrays for serial dynamic programming

… on Parallel Processing, University Park, PA, 1986

Delta transformations to simplify VLSI processor arrays for serial dynamic programming. RICHARDJ ... more Delta transformations to simplify VLSI processor arrays for serial dynamic programming. RICHARDJ LIPTON, DANIEL LOPRESTI 1986 International Conference on Parallel Processing, University Park, PA, 917-920, 1986. A ...

Research paper thumbnail of On-line handwriting recognition

Skip to Main Content. ...

Research paper thumbnail of A National Discovery Cloud: Preparing the US for Global Competitiveness in the New Era of 21st Century Digital Transformation

arXiv (Cornell University), Apr 14, 2021

Research paper thumbnail of CCC/Code 8.7: Applying AI in the Fight Against Modern Slavery

ArXiv, 2021

On any given day, tens of millions of people find themselves trapped in instances of modern slave... more On any given day, tens of millions of people find themselves trapped in instances of modern slavery. The terms"human trafficking,""trafficking in persons,"and"modern slavery"are sometimes used interchangeably to refer to both sex trafficking and forced labor. Human trafficking occurs when a trafficker compels someone to provide labor or services through the use of force, fraud, and/or coercion. The wide range of stakeholders in human trafficking presents major challenges. Direct stakeholders are law enforcement, NGOs and INGOs, businesses, local/planning government authorities, and survivors. Viewed from a very high level, all stakeholders share in a rich network of interactions that produce and consume enormous amounts of information. The problems of making efficient use of such information for the purposes of fighting trafficking while at the same time adhering to community standards of privacy and ethics are formidable. At the same time they help us,...

Research paper thumbnail of An Architecture for Programmable Systolic Arrays

Systolic algorithms are a systematic and powerful methodology for realizing the potential of very... more Systolic algorithms are a systematic and powerful methodology for realizing the potential of very large scale integration. After an initial wave of special-purpose systolic processors, research has turned to the implementation of programmable systolic machines. These machines are often too large and complicated to be able to support massive parallelism: many of the benefits to be reaped from systolic algorithms are lost. Few of the current architectures provide the simple systolic communication of the early special-purpose processors. The primary contribution of this research is the systolic shared register paradigm. Unlike previous research efforts, the architecture provides a natural means of systolic communication and uses simple processing elements; the architecture can be both easily programmed and easily formed into very large systolic arrays. The main features of the architecture are broadcast instructions, a regular topology, and register sets that are shared between neighbo...

Research paper thumbnail of Reproducible Research in Pattern Recognition, Second International Workshop, RRPR 2018

Preface This volume contains the articles from the second edition of the Workshop on Reproducible... more Preface This volume contains the articles from the second edition of the Workshop on Reproducible Research in Pattern Recognition that was held during August 20, 2018, in conjunction with ICPR 2018 in Beijing. It followed in the same spirit as the first edition with a special focus on digital geometry and mathematical morphology. It was intended as both a short participative course on reproducible research (RR) aspects leading to open discussions with the participants and also on how to actually perform RR. For this second edition, a new call for short papers was proposed to the ICPR authors. The main idea was to give authors of already accepted ICPR papers the possibility to highlight the reproducibility of their work with a companion paper. It was an opportunity to include implementation details, source code descriptions, parameter choice etc. This proceedings volume gathers 14 contributions covering the RR result track (three papers), invited RR contributions (five papers), and t...

Research paper thumbnail of An Episodic Learning Network for Text Detection on Human Bodies in Sports Images

IEEE Transactions on Circuits and Systems for Video Technology, 2021

Due to the proliferation of sports-related multimedia content on the WWW, effective visual search... more Due to the proliferation of sports-related multimedia content on the WWW, effective visual search and retrieval present interesting research challenges. These are caused by poor image quality, a wide range of possible camera points of view, pose variations on the part of athletes engaged in playing a sport, deformations of text appearing on sports person’s clothing and uniforms in motion, occlusions caused by other objects, etc. To address these challenges, this paper presents a new method for detecting text on human bodies in sports images. Unlike most existing methods, which attempt to exploit locations of a player’s torso, face, and skin, we propose an end-to-end episodic learning approach that employs inductive learning criteria for detecting clothing regions in an image, which are, in turn, then used for text detection. Our method integrates a Residual Network (ResNet) and Pyramidal Pooling Module (PPM) for generating a spatial attention map. The Progressive Scalable Expansion Algorithm (PSE) is adapted for text detection from these regions. Experimental results on our own dataset as well as several benchmarks (like RBNR and MMM which contain images of runners in marathons, and Re-ID which is a person re-identification dataset) demonstrate that the proposed method outperforms existing methods in terms of precision and F1-score. We also present results for sports images chosen from natural scene text detection datasets such as CTW1500 and MS-COCO to show the proposed method is effective and reliable across a range of inputs.

Research paper thumbnail of A New Context-Based Method for Restoring Occluded Text in Natural Scene Images

Document Analysis Systems, 2020

Research paper thumbnail of Proceedings of the 5th International Workshop on Document Analysis Systems V

Research paper thumbnail of Document Recognition V

Document Recognition V, Apr 1, 1998

Research paper thumbnail of Document Recognition and Retrieval VII

Document Recognition and Retrieval Vii, Dec 1, 1999

Research paper thumbnail of Ink Matching of Cursive Chinese Handwritten Annotations

Series in Machine Perception and Artificial Intelligence, 1999

Research paper thumbnail of Robust retrieval of noisy text

Proceedings of the Third Forum on Research and Technology Advances in Digital Libraries,

Research paper thumbnail of The Role of Document Image Analysis in Trustworthy Elections

Statistical Science and Interdisciplinary Research, 2014

Research paper thumbnail of Ink as a First-Class Datatype in Multimedia Databases

Artificial Intelligence, 1996

Research paper thumbnail of Resource-optimized delivery of web images to small-screen devices

Research paper thumbnail of <title>OCR for World Wide Web images</title>

Document Recognition IV, 1997

Research paper thumbnail of Supporting experimental research in computer vision

Research paper thumbnail of The Ninth IAPR International Workshop on Document Analysis Systems, DAS 2010, June 9-11, 2010, Boston, Massachusetts, USA

Research paper thumbnail of Discounts for dynamic programming with applications in VLSI processor arrays

Description/Abstract This dissertation introduces a method for transforming certain dynamic progr... more Description/Abstract This dissertation introduces a method for transforming certain dynamic programming problems into ones that require less space and time to solve under the logarithmic cost criterion, an appropriate complexity measure for flexible word-length ...

Research paper thumbnail of Delta transformations to simplify VLSI processor arrays for serial dynamic programming

… on Parallel Processing, University Park, PA, 1986

Delta transformations to simplify VLSI processor arrays for serial dynamic programming. RICHARDJ ... more Delta transformations to simplify VLSI processor arrays for serial dynamic programming. RICHARDJ LIPTON, DANIEL LOPRESTI 1986 International Conference on Parallel Processing, University Park, PA, 917-920, 1986. A ...