Connectionists: Postdoctoral positions at the University of Trento, Italy

[Nicu Sebe]

Multiple positions at post-doctoral level are available in the 
Department of Computer Science at the University of Trento (UniTN), to 
work under the supervision of Prof. Nicu Sebe 
(http://disi.unitn.it/~sebe/ <http://disi.unitn.it/~sebe/>) and Prof. 
Elisa Ricci (http://elisaricci.eu).

The candidates will join the MHUG (Multimedia and Human Understanding 
Group) at UniTN and be part of the Vision and Learning Joint Laboratory, 
a new laboratory bringing together computer vision researchers from 
UniTN and Fondazione Bruno Kessler (https://www.fbk.eu 
<https://www.fbk.eu/>). The research group conducts research in the 
fields of computer vision and multimedia. In computer vision, we address 
a large spectrum of themes including human-behavior analysis, action 
recognition, 2D/3D object detection, large-scale event detection and 
video analysis, etc.  In multimedia, our research focuses on cross media 
retrieval, multi-modal learning, social media analysis, emotion 
recognition, etc.

The open positions are financed by several projects: the H2020 EU 
Project SPRING (Socially Assistive Robots for Gerontological 
Healthcare), the PRIN italian project PREVUE (Prediction of Activities 
and Events by Vision in an Urban Environment), EUREGIO project OLIVER 
(Open-ended Learning for Interactive Robots), and collaboration project 
Italy-CHINA TALENT (Joint UAV and Surveillance Video Content Analysis 
and Mining for Smart City). The research group is also involved in 
several industrial projects, in collaboration with many national and 
international companies.

The main research themes are:

  *

    Theme A: Human Behavior Analysis. The activity will focus on
    designing novel algorithms and models for analyzing and
    understanding human behavior from visual data gathered by a robotic
    platform. Specifically, we expect the candidate to develop
    algorithms for behavioral cues extraction  (e.g. gestures, facial
    expressions), human action recognition/anticipation and social
    interactions analysis.

  *

    Theme B: Semantic Scene Understanding. The activity will focus on
    developing deep learning-based algorithms for semantic scene
    analysis. In particular, the research activity will focus on
    devising algorithmic solutions for pixel-level prediction tasks
    (e.g. semantic segmentation, depth estimation) and on their
    integration within tools for building semantic maps of indoor scenes.

  *

    Theme C: Domain Adaptation and Continual Learning. The activity will
    build on previous works from the research group (Mancini et al.,
    CVPR 2019; Berriel et al., ICCV 2019, Roy et al.CVPR 2019) and will
    focus on devising novel algorithms for domain adaptation and
    continual learning, with special emphasis on methodologies for video
    streams analysis.

The candidates are expected to take a lead role in the projects, 
including developing novel algorithmic solutions, setting up 
experiments, collecting and analyzing data, training and supervision of 
graduate students and undergraduate research assistants, and 
dissemination of results at conferences and in research publications.

At the time of the application, eligible candidates should have a Ph.D. 
degree in Computer Science, Engineering, or related fields. Proven 
scientific track record on major computer vision and multimedia 
conferences/journals (CVPR, ICCV, ECCV, ACM Multimedia, TPAMI, IJCV, 
etc.) is a criteria for the selection as well as experience on deep 
learning algorithms and relevant platforms (e.g. TensorFlow, PyTorch, 
Theano, Caffe). Experience in robotics is considered a plus. Desirable 
skills and qualifications are experience in grant proposal preparation 
at European and National level as well as experience in supervising or 
co-supervising PhD and MSc students.

-- 
Prof. Nicu Sebe, Ph.D.
Department of Information Engineering and Computer Science
University of Trento
Via Sommarive 9 - 38123 Povo - Trento (Italy)

ph. + 39 0461 28 2989
fax. + 39 0461 28 3939

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/connectionists/attachments/20191011/2914cabf/attachment.html>