Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
ETHypertrophy
You will use workout videos from a human study to annotate concentric and eccentric muscle contractions and total time under tension. Therefore, you can use existing computer vision and/or machine learning tools, e.g. OpenCV, Keras, PyTorch, Tensorflow.
**The Topic**
Muscle mass and strength is crucial for healthy living. Resistance exercise is an efficient and effective interventional strategy to maintain and/or increase muscle mass and strength. However, despite significant experimental attention, optimal regulation or manipulation strategies to efficiently increase muscle mass and strength are still unknown nowadays. This is mainly caused by neglecting or insufficient reporting of important mechano-biological descriptors such as e.g. time-under-tension or the temporal distribution of concentric vs eccentric contractions.
**The Project**
You will use computer vision tools and machine learning to automatically analyze 400 video sequences of people working out on resistance exercise machines where the weight stack moves up and down. You detect the movement and rest of the weight stack and reversal points automatically by using existing computer vision tools such as e.g. OpenCV, Keras, PyTorch or Tensorflow. You have the chance to become a co-author of a scientific paper in a conference or journal.
**You are**
- familiar with Java, Matlab, C/C++/C#, Python or R
- experienced with Computer Vision and OpenCV
- interested in machine learning
- capable of reading scientific articles
- motivated to participate in current research
– interested in citizen science and personal data protection
**The Topic** Muscle mass and strength is crucial for healthy living. Resistance exercise is an efficient and effective interventional strategy to maintain and/or increase muscle mass and strength. However, despite significant experimental attention, optimal regulation or manipulation strategies to efficiently increase muscle mass and strength are still unknown nowadays. This is mainly caused by neglecting or insufficient reporting of important mechano-biological descriptors such as e.g. time-under-tension or the temporal distribution of concentric vs eccentric contractions. **The Project** You will use computer vision tools and machine learning to automatically analyze 400 video sequences of people working out on resistance exercise machines where the weight stack moves up and down. You detect the movement and rest of the weight stack and reversal points automatically by using existing computer vision tools such as e.g. OpenCV, Keras, PyTorch or Tensorflow. You have the chance to become a co-author of a scientific paper in a conference or journal. **You are** - familiar with Java, Matlab, C/C++/C#, Python or R - experienced with Computer Vision and OpenCV - interested in machine learning - capable of reading scientific articles - motivated to participate in current research – interested in citizen science and personal data protection
You will build a script that accurately detects reversal points so you can calculate concentric and eccentric contraction phases and time-under-tension.
You will build a script that accurately detects reversal points so you can calculate concentric and eccentric contraction phases and time-under-tension.
Ernst Hafen, ehafen@ethz.ch, Claudio Viecelli, vclaudio@ethz.ch
Ernst Hafen, ehafen@ethz.ch, Claudio Viecelli, vclaudio@ethz.ch