I am happy to announce that the following two papers have been
accepted at the 22nd International SPIN Symposium on Model Checking of
Software (SPIN 2015) to be held in Stellenbosch, South Africa on 24–26
August 2015.
---
On Refinement of Büchi Automata for Explicit Model Checking
František Blahoudek¹, Alexandre Duret-Lutz²,
Vojtěch Rujbr¹, and Jan Strejček¹
¹Faculty of Informatics, Masaryk University, Brno, Czech Republic
²LRDE, EPITA, Le Kremlin-Bicêtre, France
https://www.lrde.epita.fr/wiki/Publications/blahoudek.15.spin
Abstract:
In explicit model checking, systems are typically described in an
implicit and compact way. Some valid information about the system
can be easily derived directly from this description, for example
that some atomic propositions cannot be valid at the same time. The
paper shows several ways to apply this information to improve the
Büchi automaton built from an LTL specification. As a result, we get
smaller automata with shorter edge labels that are easier to
understand andmore importantly, for which the explicit model
checking process performs better.
---
Practical Stutter-Invariance Checks for ω-Regular Languages
Thibaud Michaud and Alexandre Duret-Lutz
LRDE, EPITA, Le Kremlin-Bicêtre, France
https://www.lrde.epita.fr/wiki/Publications/michaud.15.spin
Abstract:
We propose several automata-based constructions that check whether a
specification is stutter-invariant. These constructions assume that
a specification and its negation can be translated into Büchi
automata, but aside from that, they are independent of the
specification formalism. These transformations were inspired by a
construction due to Holzmann and Kupferman, but that we broke down
into two operations that can have different realizations, and that
can be combined in different ways. As it turns out, implementing
only one of these operations is needed to obtain a functional
stutter-invariant check. Finally we have implemented these
techniques in a tool so that users can easily check whether an LTL
or PSL formula is stutter-invariant.
--
Alexandre Duret-Lutz
I am happy to announce that the following paper has been accepted
for publication in the Image and Vision Computing Journal (IVC):
What is a good evaluation protocol for text localization systems?
Concerns, arguments, comparisons and solutions
Stefania Calarasanu (1), Jonathan Fabrizio (1) and Séverine Dubuisson (2)
(1) LRDE-EPITA, 14-16, rue Voltaire, F-94276, Le Kremlin
Bicêtre, France
(2) CNRS, UMR 7222, ISIR, F-75005, Paris, France
Abstract:
A trustworthy protocole is essential to evaluate a text detection algorithm
in order to, first measure its efficiency and adjust its parameters and,
second to compare its performances with those of other algorithms.
However, current protocols do not give precise enough evaluations
because they use coarse evaluation metrics, and deal with inconsistent
matchings between the output of detection algorithms and the ground truth,
both often limited to rectangular shapes. In this paper, we propose a new
evaluation protocol, named EvaLTex, that solves some of the current problems
associated with classical metrics and matching strategies. Our system deals
with different kinds of annotations and detection shapes. It also considers
different kinds of granularity between detections and ground truth objects
and hence provides more realistic and accurate evaluation measures.
We use this protocol to evaluate text detection algorithms and highlight
some key examples that show that the provided scores are more relevant than
those of currently used evaluation protocols.
Ana Stefania Calarasanu
___________________________________________________
PhD Engineer
EPITA Research and Development Laboratory (LRDE)
14-16 rue Voltaire, 94276 Le Kremlin-Bicêtre CEDEX, France
https://www.lrde.epita.fr/wiki/User:Calarasanu
Ana Stefania Calarasanu
___________________________________________________
PhD Student
EPITA Research and Development Laboratory (LRDE)
14-16 rue Voltaire, 94276 Le Kremlin-Bicêtre CEDEX, France
https://www.lrde.epita.fr/wiki/User:Calarasanu
I am happy to announce that the following paper has been accepted to the
11th International Conference on Computer Vision Theory and Applications (VISAPP)
that will take place in Rome, Italy, on February 27 - 29, 2016:
Towards the rectification of highly distorted texts
Stefania Calarasanu (1), Séverine Dubuisson (2) and Jonathan Fabrizio (1)
(1) LRDE-EPITA, 14-16, rue Voltaire, F-94276, Le Kremlin
Bicêtre, France
(2) CNRS, UMR 7222, ISIR, F-75005, Paris, France
Abstract:
A frequent challenge for many Text Understanding Systems is to
tackle the variety of text characteristics in born-digital and natural
scene images to which current OCRs are not well adapted. For
example, texts in perspective are frequently present in real-word
images, but despite the ability of some detectors to accurately
localize such text objects, the recognition stage fails most of the
time. Indeed, most OCRs are not designed to handle text strings
in perspective but rather expect horizontal texts in a parallel-frontal
plane to provide a correct transcription. In this paper, we propose a
rectification procedure that can correct highly distorted texts, subject
to rotation, shearing and perspective deformations. The method is
based on an accurate estimation of the quadrangle bounding the
deformed text in order to compute a homography to transform this
quadrangle (and its content) into a horizontal rectangle.
The rectification is validated on the dataset proposed during the
ICDAR 2015 Competition on Scene Text Rectification.
Ana Stefania Calarasanu
___________________________________________________
PhD Engineer
EPITA Research and Development Laboratory (LRDE)
14-16 rue Voltaire, 94276 Le Kremlin-Bicêtre CEDEX, France
https://www.lrde.epita.fr/wiki/User:Calarasanu
Bonjour à tous,
Nous avons le plaisir de vous inviter à la soutenance de thèse d'Ana
Calarasanu intitulée ``Improvement of a text detection chain and
the proposition of a new evaluation protocol for text detection algorithms’’.
Celle-ci aura lieu le vendredi 11 décembre 2015 à 13h30 en amphi 3 à
l'EPITA, situé au 14-16 rue Voltaire au Kremlin-Bicêtre. Vous trouverez
un plan d'accès à l'école à l'adresse suivante :
https://www.lrde.epita.fr/wiki/Affiche-these-SC
La soutenance sera suivie d'un pot.
Manuscrit de thèse
------------------
Téléchargeable à cette adresse :
https://www.lrde.epita.fr/~calarasanu/manuscript_thesis_CALARASANU.pdf
Composition du jury de thèse
----------------------------
Rapporteurs :
Jean-Marc OGIER (Université La Rochelle)
Lionel PREVOST (Université des Antilles et de la Guyane)
Examinateurs :
Nicolas BREDECHE (Université Pierre et Marie Curie)
Christopher KERMORVANT (Teklia)
Beatriz MARCOTEGUI (MINES ParisTech)
Nicole VINCENT (Université Paris-Descartes)
Directeurs de thèse :
Séverine DUBUISSON (Université Pierre et Marie Curie)
Jonathan FABRIZIO (Ecole Pour l’Informatique et les Techniques Avancées)
Résumé de la thèse
------------------
The objective of this thesis is twofold. On one hand it targets
the proposition of a more accurate evaluation protocol designed
for text detection systems that solves some of the existing
problems in this area. On the other hand, it focuses on the
design of a text rectification procedure used for the correction
of highly deformed texts.
Text detection systems have gained a significant importance
during the last years. The growing number of approaches proposed
in the literature requires a rigorous performance evaluation and
ranking. In the context of text detection, an evaluation protocol
relies on three elements: a reliable text reference, a matching set
of rules deciding the relationship between the ground truth and the
detections and finally a set of metrics that produce intuitive scores.
The few existing evaluation protocols often lack accuracy either due
to inconsistent matching procedures that provide unfair scores or due
to unrepresentative metrics. Despite these issues, until today,
researchers continue to use these protocols to evaluate their work.
In this Ph.D thesis we propose a new evaluation protocol for text
detection algorithms that tackles most of the drawbacks faced by
currently used evaluation methods. This work is focused on three main
contributions: firstly, we introduce a complex text reference representation
that does not constrain text detectors to adopt a specific detection
granularity level or annotation representation; secondly, we propose a
set of matching rules capable of evaluating any type of scenario that can
occur between a text reference and a detection; and finally we show how
we can analyze a set of detection results, not only through a set of metrics,
but also through an intuitive visual representation. We use this protocol to
evaluate different text detectors and then compare the results with those
provided by alternative evaluation methods.
A frequent challenge for many Text Understanding Systems is to tackle the
variety of text characteristics in born-digital and natural scene images to
which current Optical Character Recognition (OCR)s are not well adapted.
For example, texts in perspective are frequently present in real-word images
because the camera capture angle is not normal to the plane containing text
regions. Despite the ability of some detectors to accurately localize such text
objects, the recognition stage fails most of the time. Indeed, most OCRs are
not designed to handle text strings in perspective but rather expect horizontal
texts in a parallel-frontal plane to provide a correct transcription. All these
aspects, together with the proposition of a very challenging dataset, motivated
us to propose a rectification procedure capable of correcting highly distorted texts.
