Abstract
Video segmentation is one of the most important parts of a vision system which allows partitioning each frame into homogeneous regions that share a common property. This work proposes a new methodology that aggregates three different techniques: background subtraction, region growing and a pseudo Fuzzy colour model to define colour subspaces that characterize each class. In addition, the pseudo Fuzzy colour model allows a given colour to belong to more than one class and enables the expansion ofthe classes through a dynamic model based on belonging and persistence information. In case of shared colours among classes, regional features are searched in order to determine the object's class. Tests with test and real videos of sports footages show promising results.

Abstract

Abstract
We propose an online sensorimotor architecture for controlling a low-cost humanoid robot to perform dance movements synchronized with musical stimuli. The proposed architecture attempts to overcome the robot's motor constraints by adjusting the velocity of its actuators and inter-changing the attended beat metrical-level on-the-fly. Moreover, we propose quantitative metrics for measuring the level of beat-synchrony of the generated robot dancing motion and complement them with a qualitative survey about several aspects of the demonstrated robot dance performances. Tests with different dance movements and musical pieces demonstrated satisfactory beat-synchrony results despite the physical limitations of the robot. The comparison against robot dance sequences generated without inter-changing the attended metrical-level validated our sensorimotor approach for controlling beat-synchronous robot dancing motions using different dance movements and facing distinct musical tempo conditions.

Abstract
Humanoid robots usually have a large number of degrees of freedom which turns humanoid control into a very complex problem. Humanoids are used in several RoboCup soccer leagues. In this work, the SimSpark simulator of the Simulation 3D will be used. This paper presents an automatic approach for developing humanoid behaviors based on the development of a generic optimization system. The paper describes the adopted architecture, main design choices and the results achieved with the optimization of a side kick and a forward kick. For both skills, the optimization approach allowed the creation of faster and more powerful and stable behaviors.

Abstract
The main objective of this work is to describe the Virtual Center for the Rehabilitation of Road Accident Victims - VICERAVI and the serious games it includes for the purpose of neuropsychological rehabilitation. The validation of the VICERAVI as a rehabilitation system involves 30 participants of both sexes, with brain injury due to road accidents. The applicability and the benefits of virtual worlds for a holistic neuropsychological rehabilitation will be tested using virtual reality (OpenSimulator) to simulate tasks of everyday life,,enabling participants to perform cognitive and social skills training through their avatars. The progresses of this group will se compared with a group of conventional neuropsychological rehabilitation. The main innovation of the work is the possibility if administering neuropsychological rehabilitation at distance, via ecological virtual environments. Moreover, the paper discusses also relevant criteria and issues in respect to the use of serious games, simulation and virtual environments in rehabilitation.

Abstract
This paper presents a visual localization approach that is suitable for domestic and industrial environments as it enables accurate, reliable and robust pose estimation. The mobile robot is equipped with a single camera which update sits pose whenever a landmark is available on the field of view. The innovation presented by this research focuses on the artificial landmark system which has the ability to detect the presence of the robot, since both entities communicate with each other using an infrared signal protocol modulated in frequency. Besides this communication capability, each landmark has several high intensity light-emitting diodes (LEDs) that shine only for some instances according to the communication, which makes it possible for the camera shutter and the blinking of the LEDs to synchronize. This synchronization increases the system tolerance concerning changes in brightness in the ambient lights over time, independently of the landmarks location. Therefore, the environment's ceiling is populated with several landmarks and an Extended Kalman Filter is used to combine the dead-reckoning and landmark information. This increases the flexibility of the system by reducing the number of landmarks required. The experimental evaluation was conducted in a real indoor environment with an autonomous wheelchair prototype.