Logical structure extraction of documents remains a challenging problem due to their inherent complexity and the gap between the physical features extracted from the image and their corresponding logical interpretation. Most of the literature proposes model-driven approaches which are not enough generic to handle complex and noisy documents. They do not use intermediate interpretation steps and do not explain the relationships between the physical blocks and the corresponding logical labels. The main objective of the thesis is to develop a hybrid method, using both data-driven and model-driven approach, which is capable to learn the relationships and simulate human perception during the logical recognition task. We have proposed a Dynamic P...
This paper presents a hybrid approach to learner modeling into MUSE-Logic, an ITS for logical reason...
The opaqueness of deep neural networks hinders their employment in safety-critical applications. Thi...
This thesis focuses on automatic recognition of historical French registers. These documents contain...
The original publication is available at "http://www.springerlink.com/"International audienceThis pa...
The original publication is available at www.springerlink.comInternational audienceThis paper propos...
This paper addresses the problem of layout and logical structure extraction from image documents. Tw...
The purpose of this work is to develop a pattern recognition system simulating the human vision. A t...
The paper is devoted to the description of hybrid pattern recognition method developed by research g...
Currently, most artificial neural networks (ANNs) represent relations, such as back-propagation neur...
III-formalized situations can easily be represented by means of conditional rules. Heuristics that a...
This paper describes a neural network design using auxiliary inputs, namely the indicators, that act...
The detection of semantic relationships between objects represented in an image is one of the fundam...
Artificial neural networks constitute good tools for certain types ofcomputational modeling (...
Machine reading comprehension has aroused wide concerns, since it explores the potential of model fo...
The work presented in this thesis, made in collaboration with the company Digital Scientific Researc...
This paper presents a hybrid approach to learner modeling into MUSE-Logic, an ITS for logical reason...
The opaqueness of deep neural networks hinders their employment in safety-critical applications. Thi...
This thesis focuses on automatic recognition of historical French registers. These documents contain...
The original publication is available at "http://www.springerlink.com/"International audienceThis pa...
The original publication is available at www.springerlink.comInternational audienceThis paper propos...
This paper addresses the problem of layout and logical structure extraction from image documents. Tw...
The purpose of this work is to develop a pattern recognition system simulating the human vision. A t...
The paper is devoted to the description of hybrid pattern recognition method developed by research g...
Currently, most artificial neural networks (ANNs) represent relations, such as back-propagation neur...
III-formalized situations can easily be represented by means of conditional rules. Heuristics that a...
This paper describes a neural network design using auxiliary inputs, namely the indicators, that act...
The detection of semantic relationships between objects represented in an image is one of the fundam...
Artificial neural networks constitute good tools for certain types ofcomputational modeling (...
Machine reading comprehension has aroused wide concerns, since it explores the potential of model fo...
The work presented in this thesis, made in collaboration with the company Digital Scientific Researc...
This paper presents a hybrid approach to learner modeling into MUSE-Logic, an ITS for logical reason...
The opaqueness of deep neural networks hinders their employment in safety-critical applications. Thi...
This thesis focuses on automatic recognition of historical French registers. These documents contain...