A5011001189- Handwritten Text Recognition Techniques
- Image Retrieval and Classification Techniques
- Image Processing and 3D Reconstruction
- Advanced Image and Video Retrieval Techniques
- 3D Shape Modeling and Analysis
- Robotics and Sensor-Based Localization
- Medical Image Segmentation Techniques
- 3D Surveying and Cultural Heritage
- Vehicle License Plate Recognition
- Human Pose and Action Recognition
- Image and Object Detection Techniques
- Advanced Neural Network Applications
- Face and Expression Recognition
- Natural Language Processing Techniques
- Ancient Mediterranean Archaeology and History
- Video Surveillance and Tracking Methods
- Ancient and Medieval Archaeology Studies
- Video Analysis and Summarization
- AI in cancer detection
- Computer Graphics and Visualization Techniques
- Advanced Vision and Imaging
- Remote Sensing and LiDAR Applications
- Face recognition and analysis
- Human Motion and Animation
- Autonomous Vehicle Technology and Safety
Democritus University of Thrace
2015-2024
Athena Research and Innovation Center In Information Communication & Knowledge Technologies
2016-2023
Institute of Informatics & Telecommunications
2003-2017
National Centre of Scientific Research "Demokritos"
2003-2017
Institute for Language and Speech Processing
2016
Biomechanics Institute of Valencia
2014
Institute of Informatics of the Slovak Academy of Sciences
2007
AmeriCorps
2005
University of the Aegean
2004
Institut de Recherche en Informatique et Systèmes Aléatoires
2001-2003
DIBCO 2009 is the first International Document Image Binarization Contest organized in context of ICDAR conference. The general objective contest to identify current advances document image binarization using established evaluation performance measures. This paper describes details including measures used as well 43 submitted methods along with a short description each method.
DIBCO 2013 is the international Document Image Binarization Contest organized in context of ICDAR conference. The general objective contest to identify current advances document image binarization for both machine-printed and handwritten images using evaluation performance measures that conform analysis recognition. This paper describes details including used as well 23 submitted methods along with a short description each method.
Deep convolutional neural networks (CNNs) are investigated in the context of computer-aided diagnosis (CADx) breast cancer. State-of-the-art CNNs trained and evaluated on two mammographic datasets, consisting ROIs depicting benign or malignant mass lesions. The performance evaluation each examined network is addressed training scenarios: first involves initializing with pre-trained weights, while for second initialized a random fashion. Extensive experimental results show superior achieved...
H-DIBCO 2010 is the International Document Image Binarization Contest which dedicated to handwritten document images organized in conjunction with ICFHR conference. The general objective of contest identify current advances image binarization using meaningful evaluation performance measures. This paper reports on details including measures used as well 17 submitted methods along a short description each method.
DIBCO 2011 is the International Document Image Binarization Contest organized in context of ICDAR conference. The general objective contest to identify current advances document image binarization for both machine-printed and handwritten images using evaluation performance measures that conform analysis recognition. This paper describes details including used as well 18 submitted methods along with a short description each method.
H-DIBCO 2012 is the International Document Image Binarization Competition which dedicated to handwritten document images organized in conjunction with ICFHR conference. The objective of contest identify current advances image binarization using meaningful evaluation performance measures. This paper reports on details including measures used as well 24 submitted methods along a short description each method.
Document image binarization is of great importance in the document analysis and recognition pipeline since it affects further stages process. The evaluation a method aids studying its algorithmic behavior, as well verifying effectiveness, by providing qualitative quantitative indication performance. This paper addresses pixel-based methodology for historical handwritten/machine-printed images. In proposed scheme, recall precision measures are properly modified using weighting scheme that...
A heterogeneous object (HO) modeling system independent of any commercial CAD software packages is introduced in this paper.Through system, models can be converted into 2D slices with material information for the fabrication rapid prototyping technique.In framework, volumetric dataset (VD) employed to represent variations, which offers flexible manipulability HO representation, while geometry model used describe shape an object, guarantee accuracy final slices.Two schemes are evaluate...
H-DIBCO 2016 is the international Handwritten Document Image Binarization Contest organized in context of ICFHR conference. The general objective contest to identify current advances document image binarization handwritten images using performance evaluation measures that are motivated by analysis and recognition requirements. This paper describes details including used as well 12 submitted methods along with a brief description each method.
DIBCO 2017 is the international Competition on Document Image Binarization organized in conjunction with ICDAR conference. The general objective of contest to identify current advances document image binarization machine-printed and handwritten images using performance evaluation measures that are motivated by analysis recognition requirements. This paper describes competition details including used as well 26 submitted methods along a brief description each method.
DIBCO 2019 is the international Competition on Document Image Binarization organized in conjunction with ICDAR conference. The general objective of contest to identify current advances document image binarization machine-printed and handwritten images using performance evaluation measures that are motivated by analysis recognition requirements. This paper describes competition details including used as well 24 submitted methods along a brief description each method.
Purpose An overview of the current use handwritten text recognition (HTR) on archival manuscript material, as provided by EU H2020 funded Transkribus platform. It explains HTR, demonstrates , gives examples cases, highlights affect HTR may have scholarship, and evidences this turning point advanced digitised heritage content. The paper aims to discuss these issues. Design/methodology/approach This adopts a case study approach, using development delivery one openly available platform for...
Autonomous vehicles integrate complex software stacks for realizing the necessary iterative perception, planning, and action operations. One of foundational layers such is perception one which comprised localization, detection, recognition algorithms understanding location driving environment around vehicle. More precisely, localization aims to identify vehicle on a global coordinate system considered most critical parts in stack since its accuracy robustness affects subsequent layer also...