Computer Vision & Embodied Cognition

Enabling computers and robots to interpret and understand the visual world by allowing machines to process and analyse visual data (images and videos) in the same way that humans do, extracting meaningful information from them. 

Key researchers: Jose Santos-Victor, Catarina Barata, Alexandre Bernardino, Arlindo Oliveira, Plinio Moreno

Developing advanced robots or artificial agents that mimic human-like cognitive functions and behaviours. These systems are designed to perform tasks that require reasoning, perception, learning, and decision-making, similar to human thought processes. 

Key researchers: Jose Santos-Victor, Alexandre Bernardino, Plinio Moreno

Robot vision is the capability of robots to process, interpret, and understand visual information from the world, which is similar to human vision. It involves a combination of hardware (cameras, sensors) and software (algorithms) to perform tasks like object detection, recognition, motion tracking, and scene understanding. Robots learn using ML and deep learning techniques to improve their performance over time by adapting to new data or environments.

Key researchers: Alexandre Bernardino, Plinio Moreno