Visual human-robot interaction

eye

Description


Cognitive
Emotive
Visual System


“The intelligent
glance acts in the following way:  it anticipates,
 it prevents,
it uses information already known, it
recognizes,
it interprets”

 (Jose A. Marina, ?La Tª de la
inteligencia
creadora?)

Why we call it Cognitive vision?
Because in the human vision the knowledge plays a important role, we
anticipate, we use previously acquired information, we use information
from the knowledge we have of the environment, we recognize, we
interprete. Human vision is a
intelligent vision not limited by the data obtained from the physical
reaction after a visual stimulus, it also incorporates a serie of
mechanisms that try
to assure that all the knowledge, as much as the internal (of the own
observer)
as the knowledge of the environment, is used .

Why we call it Emotive vision?
Because the affective states play a very important role in many aspects
of the human activity and above all in the interaction with others. The
fact of including emotional assessment in an artificial vision system
adds additional information that can explain behaviors that could not
be understood without the affective factor. Therefore if we want to
equip the robot with the communication skills of humans the vision
system will need to consider the incorporation of a mechanism of visual
assessment of emotions.

Entries:
Automatic Demonstration and Feature Selection for Robot Learning
IEEE International Conference on Humanoid Robots, 2015, Seoul, South Korea
S. Morante Juan G. Victores
Action Effect Generalization, Recognition and Execution through Continuous Goal-Directed Actions
IEEE International Conference on Robotics and Automation (ICRA 2014), 2014, Hong Kong, China
S. Morante Juan G. Victores A. Jardon
On Using Guided Motor Primitives to Execute Continuous Goal-Directed Actions
IEEE international symposium on robot and human interactive communication (RO-MAN 2014), 2014, Edinburgh, Scotland
S. Morante Juan G. Victores A. Jardon
Improving CGDA execution through Genetic Algorithms incorporating Spatial and Velocity constraints
IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC), 2017, Coimbra, Portugal
R. Fernandez-Fernandez D. Estévez Juan G. Victores
Reducing the Number of Evaluations Required for CGDA Execution through Particle Swarm Optimization Methods
IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC), 2017, Coimbra, Portugal
R. Fernandez-Fernandez D. Estévez Juan G. Victores
Robot Imitation through Vision, Kinesthetic and Force Features with Online Adaptation to Changing Environments
2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018, Madrid, Spain
R. Fernandez-Fernandez Juan G. Victores D. Estévez

Entries: