Marie Skłodowska-Curie Actions

Post-doctoral Research Fellowship: MSCA-IF 2017 at the Department of Computer Architecture and Technology

This hosting has expired

    01/09/2018 12:30 - Europe/Brussels
    H2020 / Marie Skłodowska-Curie Actions
    Spain, Granada
    International Project Office
    Promotion & Advisory Unit

Professor Eduardo Ros, from the Department of Computer Architecture and Technology at the University of Granada, welcomes postdoctoral candidates interested in applying for a Marie Skłodowska-Curie Individual Fellowships (MSCA-IF) in this university. Applicants must comply with the Mobility Rule (more information in the participant guide: http://sl.ugr.es/097k).

The host group is integrated within the CITIC. In the last 5 years, the CITIC was involved in 11 EU projects and one Network of Excellence within FP6, FP7 and H2020 (see http://citic.ugr.es/).

We are a leading group in the development of real-time low level vision engines for a variety of hardware platforms (CPU/GPU/FPGA), also with real-world applications in driver assistance systems, robotics, and UAVs (see http://www.ugr.es/~fbarranco/vrgroup.html). The group has also experience in bio-inspired principles such as active vision, attention, or architectural principles in low-level vision engines. The lab has collaborated with a wide set of leading researchers: University of Edinburgh (Prof. Vijayakumar, vision model abstraction), KTH Stockholm (Prof. Kragic, vision in robotics), Stanford University (Prof. Boahen, artificial retinas), Harvard University (Prof. Peli, aids for low-vision patients), Katholieke Universiteit Leuven (Prof. Van-Hulle, adaptive vision models), University of Genoa (Prof. Sabatini, low level vision based on spatio-temporal filters), Hella GmbH (Mr. Rotter, computer vision for advanced driver assistance in vehicles), or University of Maryland (Prof. Aloimonos, active vision).

With regard to real-time vision, the group has participated in EU projects such as ECOVISION, DRIVSCO, TOMSY as well as national projects. We have also hosted 3 Marie Curie projects, and our postdoctoral researchers obtained 3 Marie Curie projects in other institutions. The group has also produced a spin-off (SevenSolutions) that targets ultra-accurate timing solutions.

The research project will be focused around three main topics: High performance computing (HPC), artificial vision, neuromorphic engineering and robotics. These topics are complementary and sometimes do overlap when applied to different tasks.

High-performance horizontally applies to many different fields; particularly, it is applied to methods that require large computational resources or real-time performance (especially, for applications where closing action-perception loops is required). Our expertise on GPUs and FPGAs will help here, and could be applied to fields such as Advanced Driving assistance systems (ADAS).

Another line is focused on neuromorphic engineering using artificial retinas or frame-free event-based sensors. This topic includes all the methods, sensors, or techniques that mimic neuro-biological processes, structures, or information representations. Also, we are interested in task-driven strategies to select the most important information to prevent bandwidth bottlenecks via visual attention. Biological solutions allow for new paradigms and create opportunities of connection to other domains such as Computational Neurosciences and Cognition. At the same time, another research line of our group is focused on real robotic platforms such as the Baxter robot.

Finally, topics more related to Computer Vision are motion estimation, 3D structure from motion, visual odometry, navigation and mapping, and active vision. This know-how is essential for people interested in working with UAVs. Indeed, our group provides custom solutions, fabricating our own UAVs.

Research Area:

  • Information Science and Engineering (ENG)

For a correct evaluation of your candidature, please send the documents below to Professor (eros@ugr.es):

  • CV
  • Letter of recommendation (optional)