Motion and depth estimation for autonomous

Transcript

Motion and depth estimation for autonomous
Università degli Studi di Genova
Academic year 2015-2016
MASTER EMARO (Second Year)
European Master on Advanced Robotics
thesis short proposal
Title: Motion and depth estimation for autonomous navigation
Where: Genova
Short summary
To navigate in a real environment, e.g. to avoid obstacle and
target, is a crucial ability for artificial agents. The motion
information about the scene can be effectively computed by
inspired models of the human visual system.
The aim of this thesis is to develop an integrated neural
autonomous navigation.
The expected results are software modules (CPU or GPU-based)
and depth estimation, and their assessment in real conditions.
to reach a
and depth
using biomodel for
for motion
To better understand: 1 – 2 bibliographic references (please documents accessible
from internet)
F Solari, M Chessa, NVK Medathati, P Kornprobst. What can we expect from a
V1-MT feedforward architecture for optical flow estimation? Signal
Processing: Image Communication, 2015.
Name and references of the hosting laboratory and/or the external institution
(University, Research Centre, Private Company etc.):
University of Genoa, DIBRIS
Contacts (name, phone, email, skype..):
Manuela Chessa, +39 010 353 2289, [email protected]
Fabio Solari,
+39 010 353 2059, [email protected]