Deep neural networks and attention mechanisms for handwritten text recognition
PhD Student: Kiilian Barrère Supervisors: Bertrand Couäsnon, Aurélie Lemaitre, Yann Soullard
PhD Student: Kiilian Barrère Supervisors: Bertrand Couäsnon, Aurélie Lemaitre, Yann Soullard
PhD student: William Mocaër Supervisors: Eric ANQUETIL Richard KULPA Project: Digisport EUR
Description In the context of a work on text-lines localization in handwritten documents containing tables, we proposed to evaluate our system on a subset of cBAD (Competition on Baseline Detection, ICDAR 2017 [1]) dataset (track B) that contains exclusively documents with tabular structures. The dataset of the cBAD competition is…
Reconnaissance en-ligne d’actions 3D par l’analyse des trajectoires du squelette humain Soutenue le 11 juillet 2018 Résumé L’objectif de cette thèse est de concevoir une approche transparente originale apte à détecter en temps-réel l’occurrence d’une action, dans un flot non segmenté et idéalement le plus tôt possible. Ces travaux s’inscrivent…
Description This work takes place in the context of the thesis of of Kwon-Young Choi on the subject of resolving segmentation problems in dense and damaged printed piano scores from the 1750 to 1950 period. This data set contains 2955 images with annotations for the task of detecting or rejecting…
Presentation: Different from existing dynamic hand gesture datasets, LMDHG contains unsegmented sequences of hand gestures performed with either one hand or both hands. This dataset can therefore be used for recognition of both pre-segmented and unsegmented dynamic hand gestures using skeleton data. Download the database Download database information Content: There…
Description This work takes place in the context of the “FamilySearch” database proposed for the structure recognition competition during HIP’2013 workshop (satellite of ICDAR 2013)/ This competition focus on the localisation of 4 regions of interest inside of documents, and then a clustering in order to gather similar fields. A…
Reconnaissance et interprétation des interactions tactiles multipoints pour la composition de documents structurés Defended 28 avril 2017 Abstract Due to the popularization of the touch screen devices, nowadays people are used to conduct the human-computer interactions with touch gestures. However, limited by current studies, users can use only simple multi-touch…
Presentation: The Children Handwritings Database can be used for analysis of writing performances along a learning process based on repeats of letter writing. This database is part of IntuiScript, an innovative project that aims for designing a digital notebook dedicated to handwriting learning at primary schools. One of the main…