<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
</head>
<body>
<div aria-live="assertive" id="magicdomid229" class="ace-line"><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">The
FOX team from the CRIStAL laboratory (UMR CNRS), Lille France is
looking to recruit a PhD student starting on <b class="">October
1st 2022</b> on the following subject : </span><span class="b
author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt"><b
class="">Spatio-temporal data augmentation models for motion
pattern learning using deep learning: applications to facial
analysis in the wild</b></span></div>
<div aria-live="assertive" id="magicdomid41" class="ace-line"><br
class="">
</div>
<div aria-live="assertive" id="magicdomid42" class="ace-line"><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">The
FOX research group is part of the CRIStAL laboratory (University
of Lille, CNRS), located in Lille, France. We focus on video
analysis for human behavior understanding. Specifically, we
develop spatio-temporal models of motions for tasks such as
abnormal event detection, emotion recognition, and face
alignment. Our work is published in major journals (Pattern
Recognition, IEEE Trans. on Affective Computing) and conferences
(WACV, IJCNN).<br>
<br>
</span>
<div aria-live="assertive" id="magicdomid59" class="ace-line"><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">This
PHD thesis will be funded in the framework of the </span><span
class="b author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt"><b
class="">AI_PhD@Lille</b></span><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">program.</span></div>
<div aria-live="assertive" id="magicdomid60" class="ace-line"><span
class="url
author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt"><a
href="http://www.isite-ulne.fr/index.php/en/phd-in-artificial-intelligence/"
rel="noreferrer noopener" class="moz-txt-link-freetext">http://www.isite-ulne.fr/index.php/en/phd-in-artificial-intelligence/</a></span></div>
<div aria-live="assertive" id="magicdomid61" class="ace-line"><span
class="url
author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt"><br
class="">
</span></div>
<div aria-live="assertive" id="magicdomid63" class="ace-line"><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">The
candidate will be funded for 3 years; he/she is expected to
defend his/her thesis and graduate by the end of the contract.
The monthly net salary is around <b>1800</b>€, including
benefits (health insurance, retirement fund, and paid
vacations). </span><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt"></span></div>
<div aria-live="assertive" id="magicdomid64" class="ace-line"><br
class="">
</div>
<div aria-live="assertive" id="magicdomid65" class="ace-line"><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">The
position is located in </span><span class="b
author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt"><b
class="">Lille, France</b></span><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">.
With over 110 000 students, the metropolitan area of Lille is
one France's top education student cities. The European
Doctoral College Lille Nord-Pas de Calais is headquartered in
Lille Metropole and includes 3,000 PhD Doctorate students
supported by university research laboratories. Lille has a
convenient location in the European high-speed rail network.
It lies on the Eurostar line to London (1:20 hour journey).
The French TGV network also puts it only 1 hour from Paris, 35
mn from Brussels, and a short trips to other major centres in
France such as Paris, Marseille and Lyon.</span></div>
<div aria-live="assertive" id="magicdomid66" class="ace-line"><br
class="">
</div>
<span class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt"></span></div>
<div aria-live="assertive" id="magicdomid43" class="ace-line"><br
class="">
</div>
<div aria-live="assertive" id="magicdomid44" class="ace-line"><span
class="b author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt"><b
class="">Abstract</b></span><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">:
Facial expression analysis is a well-studied field when dealing
with segmented and constrained data captured in lab conditions.
However, many challenges must still be addressed for building
in-the-wild solutions that account for various motion
intensities, strong head movements during expressions, the
spotting of the subsequence containing the expression, partially
occluded faces, etc. In recent years, learned features based on
deep learning architectures were proposed in order to deal with
these challenges. Deep learning is characterized by neural
architectures that depend on a huge number of parameters. The
convergence of these neural networks and the estimation of
optimal parameters require large amounts of training data,
especially when dealing with spatio-temporal data, particulary
adequate for facial expression recognition. The quantity, but
also the quality, of the data and its capacity to reflect the
addressed challenges are key elements for training properly the
networks. Augmenting the data artificially in an intelligent and
controlled way is an interesting solution. The augmentation
techniques identified in the literature are mainly focused on
image augmentation and consist of scaling, rotation, and
flipping operations, or they make use of more complex
adversarial training. These techniques can be applied at the
frame level, but there is a need for sequence level augmentation
in order to better control the augmentation process and ensure
the absence of temporal artifacts that might bias the learning
process. The generation of dynamic frontal facial expressions
has already been addressed in the literature. The goal of this
Ph.D. is to conceive new space-time augmentation methods for
unconstrained facial analysis (involving head movements,
occultations, etc.). Attention should be paid in assessing the
quality standards related to facial expression requirements:
stability over time, absence of facial artifacts, etc. More
specifically, the Ph.D. can</span><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">didate
is expected to conceive augmentation architectures that address
various challenges (motion intensities, head movements) while
maintaining temporal stability and eliminating facial artifacts.</span>
</div>
<div aria-live="assertive" id="magicdomid46" class="ace-line"><br
class="">
</div>
<div aria-live="assertive" id="magicdomid47" class="ace-line"><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">More
details are available here : </span><span class="url
author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt"><a
href="http://bit.ly/st_augm_motion" rel="noreferrer noopener"
class="moz-txt-link-freetext">https://bit.ly/staugm_motion</a></span></div>
<div aria-live="assertive" id="magicdomid48" class="ace-line"><br
class="">
</div>
<div aria-live="assertive" id="magicdomid49" class="ace-line"><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">Candidates
must hold a Master degree in Computer Science, Statistics,
Applied Mathematics or a related field. Experience in one or
more of the following is a plus:</span></div>
<div aria-live="assertive" id="magicdomid50" class="ace-line"><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">•
image processing, computer vision;</span></div>
<div aria-live="assertive" id="magicdomid51" class="ace-line"><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">•
machine learning;</span></div>
<div aria-live="assertive" id="magicdomid52" class="ace-line"><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">•
research methodology (literature review, experimentation…).</span></div>
<div aria-live="assertive" id="magicdomid53" class="ace-line"><br
class="">
</div>
<div aria-live="assertive" id="magicdomid54" class="ace-line"><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">Candidates
should have the following skills:</span></div>
<div aria-live="assertive" id="magicdomid55" class="ace-line"><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">•
good proficiency in English, both spoken and written;</span></div>
<div aria-live="assertive" id="magicdomid56" class="ace-line"><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">•
scientific writing;</span></div>
<div aria-live="assertive" id="magicdomid57" class="ace-line"><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">•
programming (experience in C++ is a plus, but not mandatory).</span></div>
<div aria-live="assertive" id="magicdomid58" class="ace-line"><br
class="">
</div>
<span class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt">We
look forward to receiving your application</span><span class="i
author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt"><i class="">
as soon as possible</i></span><span
class="author-a-hdz81zz70zkz84z8z69zygz68zz70zz70zz70zz80zt"></span>
</body>
</html>