Cookies Policy
The website need some cookies and similar means to function. If you permit us, we will use those means to collect data on your visits for aggregated statistics to improve our service. Find out More
Accept Reject
  • Menu
Research Opportunities
Apply now Final Selection Minute View Formal Call
Research Opportunities

Computer Vision

[Closed]

Work description

Human pose estimation is the task that seeks to find the position and orientation of joints in a person's body, for example, in a single frame or a sequence of images. It plays an important role in understanding human movement, having wide-ranging applications in domains such as robotics, human-computer interaction and healthcare. By accurately estimating human pose, it becomes possible to recognize gestures, facilitate immersive games, improve rehabilitation and physical therapy, and enable more natural and intuitive human-machine interactions. For pose detection, wearable, pressure and vision-based sensors are used. However, wearable sensors can cause discomfort during exercise and can induce unnatural movements that lead to incorrect postures, while pressure sensors only allow the evaluation of a reduced number of exercises. On the other hand, vision-based approaches without markers do not interfere with the patient and allow a wide range of exercises to be captured. Furthermore, using a deep learning approach can capture the most significant features, leading to highly accurate human pose estimates. Taking this into consideration, the objective of the work to be developed within the scope of this grant is the exploration of deep learning (DL) approaches for the detection/tracking of the upper limb and its joints based on vision sensors. The identified frameworks must be analyzed, tested and their performance compared according to a set of key performance indicators (KPI), to be defined.

Academic Qualifications

Attendance of a Master's degree in Biomedical Engineering, Master's degree in Bioengineering, or related areas. The scholarship award assumes that the candidate is a student of a study cycle or a non-degree course taught at a Higher Education Institution.

Minimum profile required

The candidate must be enrolled in a Master's degree in Biomedical Engineering, Master's in Bioengineering, or related areas.Knowledge of programming in Python.

Preference factors

Knowledge of Artificial Intelligence, in particular Generative Adversarial Network (GAN). Knowledge of open source pose detection frameworks.

Application Period

Since 24 Nov 2023 to 11 Dec 2023

[Closed]

Centre

Robotics in Industry and Intelligent Systems

Scientific Advisor

Manuel Santos Silva