Assessment of a novel deep learning-based marker-less motion capture system for gait study

Vafadar, Saman; Skalli, Wafa; Bonnet-Lebrun, Aurore; Assi, Ayman; Gajny, Laurent

doi:10.1016/j.gaitpost.2022.03.008

Cited by 17 publications

(17 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Vafadar, et al 11 employed a pose estimation algorithm 34 that was first trained on the marker-based Human 3.6M open access dataset 20 , which was then refined using transfer learning on the ENSAM clinical dataset 11 . The ENSAM dataset is of particular interest as it combines 3D marker-based motion capture, markerless video and biplanar x-ray (EOS), to ensuring accurate identification of joint centre locations 11,35 . Vafadar, et al 35 observed significantly improved results after refinement, with considerably decreased systematic differences.…”

Section: Discussionmentioning

confidence: 99%

Examination of 2D frontal and sagittal markerless motion capture: Implications for 2D and 3D markerless applications

Wade

Needham

Evans

et al. 2023

Preprint

View full text Add to dashboard Cite

This study examined if occluded joint locations from markerless motion capture produced 2D joint angles with reduced accuracy compared to visible joints, and if 2D frontal plane joint angles were usable for practical applications. Fifteen healthy participants performed over-ground walking whilst recorded by fifteen marker-based cameras and two machine vision cameras (frontal and sagittal plane). Repeated measures Bland-Altman analysis illustrated that markerless standard deviation of bias (random differences) for the occluded-side hip and knee joint angles in the sagittal plane were double that of the camera-side (visible) hip and knee. Camera-side sagittal plane knee and hip angles were near or within marker-based error values previously observed. While frontal plane random differences accounted for 35-46% of total range of motion at the hip and knee, systematic and random differences (-4.6-1.6 +/- 3.7-4.2 degrees) were actually similar to previously reported marker-based error values. This was not true for the ankle, where random difference (+/- 12 degrees) was still too high for practical applications. Our results add to previous literature, highlighting shortcomings of current pose estimation algorithms and labelled datasets. As such, this paper finishes by reviewing marker-based methods for creating anatomically accurate markerless training data.

show abstract

Section: Discussionmentioning

confidence: 99%

Examination of 2D frontal and sagittal markerless motion capture: Implications for 2D and 3D markerless applications

Wade

Needham

Evans

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…In the training set, 14 subjects were asymptomatic adults (≥18 years), one adult had scoliosis, and 12 minors (<18 years) suffered from X-linked hypophosphatemia (XLH) disease. While in the test set, 8 adults were asymptomatic, one adult had spondylolisthesis, and 7 children had XLH [ 5 , 6 ]. The dataset comprised a total of 120,293 frames, each containing four images from four calibrated and synchronized cameras (GoPro Hero 7 Black).…”

Section: Methodsmentioning

confidence: 99%

“…Using the 17 coordinates, joint angles were computed for the lower and upper extremities. For the joint angles of the lower limbs, we followed the calculation method proposed in [ 6 ]. For the upper limbs, the approach employed in [ 27 ] was modified to fit the human model defined in this study, and we applied it to establish the local coordinate system of the segment and calculate the corresponding joint angles (see Figure 4 ).…”

Section: Methodsmentioning

confidence: 99%

“…The focus of human pose estimation is the calculation of human body keypoint coordinates based on images. Combined with kinematic analysis, it has the potential for many applications in different fields, e.g., ergonomics [ 1 , 2 , 3 , 4 ] or orthopedics [ 5 , 6 ]. In vision-based human pose estimation tasks, the datasets used for training and testing models often consist of images where the human face is clearly visible.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Effect of Face Blurring on Human Pose Estimation: Ensuring Subject Privacy for Medical and Occupational Health Applications

Jiang

Skalli

Siadat

et al. 2022

Sensors

Self Cite

View full text Add to dashboard Cite

The face blurring of images plays a key role in protecting privacy. However, in computer vision, especially for the human pose estimation task, machine-learning models are currently trained, validated, and tested on original datasets without face blurring. Additionally, the accuracy of human pose estimation is of great importance for kinematic analysis. This analysis is relevant in areas such as occupational safety and clinical gait analysis where privacy is crucial. Therefore, in this study, we explore the impact of face blurring on human pose estimation and the subsequent kinematic analysis. Firstly, we blurred the subjects’ heads in the image dataset. Then we trained our neural networks using the face-blurred and the original unblurred dataset. Subsequently, the performances of the different models, in terms of landmark localization and joint angles, were estimated on blurred and unblurred testing data. Finally, we examined the statistical significance of the effect of face blurring on the kinematic analysis along with the strength of the effect. Our results reveal that the strength of the effect of face blurring was low and within acceptable limits (<1°). We have thus shown that for human pose estimation, face blurring guarantees subject privacy while not degrading the prediction performance of a deep learning model.

show abstract

“…In this process, the 3D skeleton is meant to correspond to the model. Only when the correspondence is good, can we lay a good foundation for "skinning" [6]. Te main function of skinning is to make the bones and the model closely connected to form a whole, making it look more harmonious and unifed.…”

Section: Introductionmentioning

confidence: 99%

Research on Dance Motion Capture Technology for Visualization Requirements

Sun

2022

Scientific Programming

View full text Add to dashboard Cite

“Motion capture technology” refers to the use of optical sensing equipment to record the dancer’s movement trajectory during dance and then convert this movement trajectory into applicable data information in animation software. In the 3D animation software, the corresponding dancer model can be constructed, and the matching costumes can also be designed. After combining it with the motion capture data, animation and dance data can be generated. In this way, virtual simulation software can create some virtual visualization scenes and present more diverse and complex demonstration effects. This article focuses on three aspects: “an overview of the connotation of motion capture technology,” “design of virtual dance visualization scene based on motion capture technology,” and “application of virtual dance visualization scene based on motion capture technology.”

show abstract

Assessment of a novel deep learning-based marker-less motion capture system for gait study

Cited by 17 publications

References 28 publications

Examination of 2D frontal and sagittal markerless motion capture: Implications for 2D and 3D markerless applications

Examination of 2D frontal and sagittal markerless motion capture: Implications for 2D and 3D markerless applications

Effect of Face Blurring on Human Pose Estimation: Ensuring Subject Privacy for Medical and Occupational Health Applications

Research on Dance Motion Capture Technology for Visualization Requirements

Contact Info

Product

Resources

About