The development and evaluation of a fully automated markerless motion capture workflow.

This study presented a fully automated deep learning based markerless motion capture workflow and evaluated its performance against marker-based motion capture during overground running, walking and counter movement jumping. Multi-view high speed (200 Hz) image data were collected concurrently with marker-based motion capture (criterion data), permitting a direct comparison between methods. Lower limb kinematic data for 15 participants were computed using 2D pose estimation, our 3D fusion process and OpenSim based inverse kinematics modelling. Results demonstrated high levels of agreement for lower limb joint angles, with mean differences ranging "0.1° - 10.5° for hip (3 DoF) joint rotations, and 0.7° - 3.9° for knee (1 DoF) and ankle (2 DoF) rotations. These differences generally fall within the documented uncertainties of marker-based motion capture, suggesting that our markerless approach could be used for appropriate biomechanics applications. We used an open-source, modular and customisable workflow, allowing for integration with other popular biomechanics tools such as OpenSim. By developing open-source tools, we hope to facilitate the democratisation of markerless motion capture technology and encourage the transparent development of markerless methods. This presents exciting opportunities for biomechanics researchers and practitioners to capture large amounts of high quality, ecologically valid data both in the laboratory and in the wild.

Development, evaluation and application of a novel markerless motion analysis system to understand push-start technique in elite skeleton athletes.

This study describes the development, evaluation and application of a computer vision and deep learning system capable of capturing sprinting and skeleton push start step characteristics and mass centre velocities (sled and athlete). Movement data were captured concurrently by a marker-based motion capture system and a custom markerless system. High levels of agreement were found between systems, particularly for spatial based variables (step length error 0.001 ± 0.012 m) while errors for temporal variables (ground contact time and flight time) were on average within ± 1.5 frames of the criterion measures. Comparisons of sprinting and pushing revealed decreased mass centre velocities as a result of pushing the sled but step characteristics were comparable to sprinting when aligned as a function of step velocity. There were large asymmetries between the inside and outside leg during pushing (e.g. 0.22 m mean step length asymmetry) which were not present during sprinting (0.01 m step length asymmetry). The observed asymmetries suggested that force production capabilities during ground contact were compromised for the outside leg. The computer vision based methods tested in this research provide a viable alternative to marker-based motion capture systems. Furthermore, they can be deployed into challenging, real world environments to non-invasively capture data where traditional approaches are infeasible.

Using genetic algorithms to uncover individual differences in how humans represent facial emotion.

Emotional facial expressions critically impact social interactions and cognition. However, emotion research to date has generally relied on the assumption that people represent categorical emotions in the same way, using standardized stimulus sets and overlooking important individual differences. To resolve this problem, we developed and tested a task using genetic algorithms to derive assumption-free, participant-generated emotional expressions. One hundred and five participants generated a subjective representation of happy, angry, fearful and sad faces. Population-level consistency was observed for happy faces, but fearful and sad faces showed a high degree of variability. High test-retest reliability was observed across all emotions. A separate group of 108 individuals accurately identified happy and angry faces from the first study, while fearful and sad faces were commonly misidentified. These findings are an important first step towards understanding individual differences in emotion representation, with the potential to reconceptualize the way we study atypical emotion processing in future research.

The accuracy of several pose estimation methods for 3D joint centre localisation.

Human movement researchers are often restricted to laboratory environments and data capture techniques that are time and/or resource intensive. Markerless pose estimation algorithms show great potential to facilitate large scale movement studies 'in the wild', i.e., outside of the constraints imposed by marker-based motion capture. However, the accuracy of such algorithms has not yet been fully evaluated. We computed 3D joint centre locations using several pre-trained deep-learning based pose estimation methods (OpenPose, AlphaPose, DeepLabCut) and compared to marker-based motion capture. Participants performed walking, running and jumping activities while marker-based motion capture data and multi-camera high speed images (200 Hz) were captured. The pose estimation algorithms were applied to 2D image data and 3D joint centre locations were reconstructed. Pose estimation derived joint centres demonstrated systematic differences at the hip and knee (~ 30-50 mm), most likely due to mislabeling of ground truth data in the training datasets. Where systematic differences were lower, e.g., the ankle, differences of 1-15 mm were observed depending on the activity. Markerless motion capture represents a highly promising emerging technology that could free movement scientists from laboratory environments but 3D joint centre locations are not yet consistently comparable to marker-based motion capture.

Can Markerless Pose Estimation Algorithms Estimate 3D Mass Centre Positions and Velocities during Linear Sprinting Activities?

The ability to accurately and non-invasively measure 3D mass centre positions and their derivatives can provide rich insight into the physical demands of sports training and competition. This study examines a method for non-invasively measuring mass centre velocities using markerless human pose estimation and Kalman smoothing. Marker (Qualysis) and markerless (OpenPose) motion capture data were captured synchronously for sprinting and skeleton push starts. Mass centre positions and velocities derived from raw markerless pose estimation data contained large errors for both sprinting and skeleton pushing (mean ± SD = 0.127 ± 0.943 and -0.197 ± 1.549 m·s-1, respectively). Signal processing methods such as Kalman smoothing substantially reduced the mean error (±SD) in horizontal mass centre velocities (0.041 ± 0.257 m·s-1) during sprinting but the precision remained poor. Applying pose estimation to activities which exhibit unusual body poses (e.g., skeleton pushing) appears to elicit more erroneous results due to poor performance of the pose estimation algorithm. Researchers and practitioners should apply these methods with caution to activities beyond sprinting as pose estimation algorithms may not generalise well to the activity of interest. Retraining the model using activity specific data to produce more specialised networks is therefore recommended.

A Review of the Evolution of Vision-Based Motion Analysis and the Integration of Advanced Computer Vision Methods Towards Developing a Markerless System.

BACKGROUND: The study of human movement within sports biomechanics and rehabilitation settings has made considerable progress over recent decades. However, developing a motion analysis system that collects accurate kinematic data in a timely, unobtrusive and externally valid manner remains an open challenge. MAIN BODY: This narrative review considers the evolution of methods for extracting kinematic information from images, observing how technology has progressed from laborious manual approaches to optoelectronic marker-based systems. The motion analysis systems which are currently most widely used in sports biomechanics and rehabilitation do not allow kinematic data to be collected automatically without the attachment of markers, controlled conditions and/or extensive processing times. These limitations can obstruct the routine use of motion capture in normal training or rehabilitation environments, and there is a clear desire for the development of automatic markerless systems. Such technology is emerging, often driven by the needs of the entertainment industry, and utilising many of the latest trends in computer vision and machine learning. However, the accuracy and practicality of these systems has yet to be fully scrutinised, meaning such markerless systems are not currently in widespread use within biomechanics. CONCLUSIONS: This review aims to introduce the key state-of-the-art in markerless motion capture research from computer vision that is likely to have a future impact in biomechanics, while considering the challenges with accuracy and robustness that are yet to be addressed.