Joint constraint systems are essential components of anatomical models, ensuring the correct movement of limbs during simulation. In computer vision applications anatomical models (or priors) are fitted to recovered image or motion capture data. Anatomical models are also used in biomechanical simulations to model patients and in ergonomics modelling average individuals. Dynamics-based models are increasingly used in biomechanics and rehabilitation research in bimodal approaches combining biomechanical models with experimental data (e.g. measured external forces) .
Papers:
Y. Jiang and C. K. Liu, “Data-driven approach to simulating realistic human joint constraints,” in IEEE International Conference on Robotics and Automation (ICRA), vol. abs/1709.08685, May 2018.
G. L. Jenkins, G. Roger, M. E. Dacey, and T. Bashford, “A rigid map neural network for anatomical joint constraint modelling,” in 18th International UKSim Conference on Computer Modelling and Simulation, Cambridge, UK, April 2016.
L. Zhu, X. Hu, and L. Kavan, “Adaptable anatomical models for realistic bone motion reconstruction,” in Computer Graphics Forum, vol. 34, no. 2. Wiley Online Library, 2015, pp. 459–471.
Eduard Babulak can you provide a little more detail. I've looked at a number of neural network techniques for anatomical modelling. I'm not sure whether you mean a standard metric for comparison of approaches (in unit quaternion space,) or for evaluating their application.
- Badges
- Science topic
- Similar topics
- Correction