Towards computational modelling of active response in cyclist falls from in-the-wild footage
(2022) ESB 2022 p.1-1- Abstract
- The self-protective effects of active response in unplanned falls is a well-established, yet under-researched area. Our recent study demonstrated how single cyclist falls are prevalent (i.e., cyclist road traffic collisions not involving another road user), and that active response is of particular importance in these cases. This phenomenon has not been considered in previous computational modelling studies involving passive multibody models. Multibody dynamics solvers implement the equations of motion as an initial value problem, and it is known that resulting impact kinematics and dynamics are influenced significantly by initial posture and motion. In recent years, many deep learning and computer vision-based 3D human pose estimation... (More)
- The self-protective effects of active response in unplanned falls is a well-established, yet under-researched area. Our recent study demonstrated how single cyclist falls are prevalent (i.e., cyclist road traffic collisions not involving another road user), and that active response is of particular importance in these cases. This phenomenon has not been considered in previous computational modelling studies involving passive multibody models. Multibody dynamics solvers implement the equations of motion as an initial value problem, and it is known that resulting impact kinematics and dynamics are influenced significantly by initial posture and motion. In recent years, many deep learning and computer vision-based 3D human pose estimation methods have been developed. These techniques allow for inference of joint positions from monocular/stereo camera footage. In this study, we develop a pipeline for applying joint degrees of freedom (DOFs) and their 1st derivatives (linear and angular velocity) to represent active response using Madymo’s pedestrian multibody ellipsoid model. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/344bcb90-ba5a-499a-9d55-8b127ec00e34
- author
- Gildea, Kevin LU and Simms, Ciaran
- publishing date
- 2022
- type
- Contribution to conference
- publication status
- published
- subject
- keywords
- Computational modelling, Injury biomechanics, Human pose estimation, Deep learning, Computer vision, Video analysis
- pages
- 1 - 1
- conference name
- ESB 2022
- conference location
- Porto, Portugal
- conference dates
- 2022-06-26 - 2022-06-29
- language
- English
- LU publication?
- no
- id
- 344bcb90-ba5a-499a-9d55-8b127ec00e34
- date added to LUP
- 2022-12-13 20:07:01
- date last changed
- 2023-06-09 11:55:00
@misc{344bcb90-ba5a-499a-9d55-8b127ec00e34, abstract = {{The self-protective effects of active response in unplanned falls is a well-established, yet under-researched area. Our recent study demonstrated how single cyclist falls are prevalent (i.e., cyclist road traffic collisions not involving another road user), and that active response is of particular importance in these cases. This phenomenon has not been considered in previous computational modelling studies involving passive multibody models. Multibody dynamics solvers implement the equations of motion as an initial value problem, and it is known that resulting impact kinematics and dynamics are influenced significantly by initial posture and motion. In recent years, many deep learning and computer vision-based 3D human pose estimation methods have been developed. These techniques allow for inference of joint positions from monocular/stereo camera footage. In this study, we develop a pipeline for applying joint degrees of freedom (DOFs) and their 1st derivatives (linear and angular velocity) to represent active response using Madymo’s pedestrian multibody ellipsoid model.}}, author = {{Gildea, Kevin and Simms, Ciaran}}, keywords = {{Computational modelling; Injury biomechanics; Human pose estimation; Deep learning; Computer vision; Video analysis}}, language = {{eng}}, pages = {{1--1}}, title = {{Towards computational modelling of active response in cyclist falls from in-the-wild footage}}, url = {{https://lup.lub.lu.se/search/files/150139814/Injury_Biomechanics.pdf}}, year = {{2022}}, }