Search Results for author: Thierry Paquet

Found 21 papers, 10 papers with code

DANIEL: A fast Document Attention Network for Information Extraction and Labelling of handwritten documents

1 code implementation12 Jul 2024 Thomas Constum, Pierrick Tranouez, Thierry Paquet

Despite this, these integrated approaches have not yet matched the performance of language models, when applied to information extraction in plain text.

Document Layout Analysis document understanding +6

End-to-end information extraction in handwritten documents: Understanding Paris marriage records from 1880 to 1940

no code implementations30 Apr 2024 Thomas Constum, Lucas Preel, Théo Larcher, Pierrick Tranouez, Thierry Paquet, Sandra Brée

The EXO-POPP project aims to establish a comprehensive database comprising 300, 000 marriage records from Paris and its suburbs, spanning the years 1880 to 1940, which are preserved in over 130, 000 scans of double pages.

Handwritten Text Recognition

Sheet Music Transformer: End-To-End Optical Music Recognition Beyond Monophonic Transcription

1 code implementation12 Feb 2024 Antonio Ríos-Vila, Jorge Calvo-Zaragoza, Thierry Paquet

State-of-the-art end-to-end Optical Music Recognition (OMR) has, to date, primarily been carried out using monophonic transcription techniques to handle complex score layouts, such as polyphony, often by resorting to simplifications or specific adaptations.

Faster DAN: Multi-target Queries with Document Positional Encoding for End-to-end Handwritten Document Recognition

1 code implementation25 Jan 2023 Denis Coquenet, Clément Chatelain, Thierry Paquet

Recent advances in handwritten text recognition enabled to recognize whole documents in an end-to-end way: the Document Attention Network (DAN) recognizes the characters one after the other through an attention-based prediction process until reaching the end of the document.

Handwritten Document Recognition

Stochastic gradient descent with gradient estimator for categorical features

1 code implementation8 Sep 2022 Paul Peseux, Maxime Berar, Thierry Paquet, Victor Nicollet

Categorical data are present in key areas such as health or supply chain, and this data require specific treatment.

Confidence Estimation for Object Detection in Document Images

no code implementations29 Aug 2022 Mélodie Boillet, Christopher Kermorvant, Thierry Paquet

In the active learning framework, the three first estimators show a significant improvement in performance for the detection of document physical pages and text lines compared to a random selection of images.

Active Learning Descriptive +3

DAN: a Segmentation-free Document Attention Network for Handwritten Document Recognition

1 code implementation23 Mar 2022 Denis Coquenet, Clément Chatelain, Thierry Paquet

For the first time, we propose an end-to-end segmentation-free architecture for the task of handwritten document recognition: the Document Attention Network.

Decoder Handwritten Document Recognition +1

Robust Text Line Detection in Historical Documents: Learning and Evaluation Methods

no code implementations23 Mar 2022 Mélodie Boillet, Christopher Kermorvant, Thierry Paquet

We present a study conducted using three state-of-the-art systems Doc-UFCN, dhSegment and ARU-Net and show that it is possible to build generic models trained on a wide variety of historical document datasets that can correctly segment diverse unseen pages.

document understanding Line Detection +1

Including Keyword Position in Image-based Models for Act Segmentation of Historical Registers

no code implementations17 Sep 2021 Mélodie Boillet, Martin Maarand, Thierry Paquet, Christopher Kermorvant

However, the segmentation of complex documents into semantic regions is sometimes impossible relying only on visual features and recent models embed both visual and textual information.

Position

Have convolutions already made recurrence obsolete for unconstrained handwritten text recognition ?

no code implementations9 Dec 2020 Denis Coquenet, Yann Soullard, Clément Chatelain, Thierry Paquet

This has a direct influence on the training time of such architectures, with also a direct consequence on the time required to explore various architectures.

Handwriting Recognition Handwritten Text Recognition

End-to-end Handwritten Paragraph Text Recognition Using a Vertical Attention Network

1 code implementation7 Dec 2020 Denis Coquenet, Clément Chatelain, Thierry Paquet

For each text line features, a decoder module recognizes the character sequence associated, leading to the recognition of a whole paragraph.

Decoder Handwritten Text Recognition

CTCModel: a Keras Model for Connectionist Temporal Classification

5 code implementations23 Jan 2019 Yann Soullard, Cyprien Ruffino, Thierry Paquet

We report an extension of a Keras Model, called CTCModel, to perform the Connectionist Temporal Classification (CTC) in a transparent way.

Classification General Classification

A Unified Multilingual Handwriting Recognition System using multigrams sub-lexical units

no code implementations28 Aug 2018 Wassim Swaileh, Yann Soullard, Thierry Paquet

This makes pos- sible the design of an end-to-end unified multilingual recognition system where both a single optical model and a single language model are trained on all the languages.

Handwriting Recognition Language Modelling +1

A syllable based model for handwriting recognition

no code implementations22 Aug 2018 Wassim Swaileh, Thierry Paquet

In this paper, we introduce a new modeling approach of texts for handwriting recognition based on syllables.

Handwriting Recognition

LV-ROVER: Lexicon Verified Recognizer Output Voting Error Reduction

no code implementations24 Jul 2017 Bruno Stuner, Clément Chatelain, Thierry Paquet

Offline handwritten text line recognition is a hard task that requires both an efficient optical character recognizer and language model.

Handwriting Recognition Language Modelling

Handwriting recognition using Cohort of LSTM and lexicon verification with extremely large lexicon

no code implementations22 Dec 2016 Bruno Stuner, Clément Chatelain, Thierry Paquet

State-of-the-art methods for handwriting recognition are based on Long Short Term Memory (LSTM) recurrent neural networks (RNN), which now provides very impressive character recognition performance.

Handwriting Recognition

Logical segmentation for article extraction in digitized old newspapers

no code implementations3 Oct 2012 Thomas Palfray, David Hébert, Stéphane Nicolas, Pierrick Tranouez, Thierry Paquet

Our back-end system extracts the logical structure of the page to produce the informative units: the articles.

Retrieval

Cannot find the paper you are looking for? You can Submit a new open access paper.