Applications of Pose Estimation in Human Health and Performance across the Lifespan

Stenum, Jan; Cherry‐Allen, Kendra M.; Pyles, Connor; Reetzke, Rachel; Vignos, Michael F.; Roemmich, Ryan T.

doi:10.3390/s21217315

Cited by 85 publications

(48 citation statements)

References 126 publications

(167 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The applications of human pose estimation include some areas such as activity recognition, motion capture and augmented reality, training robots, and motion tracking for consoles [ 3 , 57 ]. Stenum et al [ 1 ] developed an application that evaluates human body performance over the lifespan based on human pose estimation. At the same time, the authors also analyzed the challenges and limitations of human-posture-based applications as the problems of hidden body parts, limited training data, limited capture errors, limited positional errors, and limited recording devices.…”

Section: Related Workmentioning

confidence: 99%

“… Illustrating of using human posture estimation in weightlifting practice ( a ) [ 5 ], healthcare, sports ( b , d ) [ 1 ], and robotics ( c ) [ 3 ]. …”

Section: Figurementioning

confidence: 99%

“…Human pose estimation is regarded as one of the most interesting research areas in computer vision. It is applied to many fields such as healthcare, sports [ 1 ], activity recognition [ 2 ], motion capture and augmented reality, training robots, motion tracking for consoles [ 3 ], etc. Barla [ 4 ] presented seven applications of human pose estimation.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Unified End-to-End YOLOv5-HR-TCM Framework for Automatic 2D/3D Human Pose Estimation for Real-Time Applications

Nguyen¹,

Nguyen²,

Scherer

et al. 2022

Sensors

View full text Add to dashboard Cite

Three-dimensional human pose estimation is widely applied in sports, robotics, and healthcare. In the past five years, the number of CNN-based studies for 3D human pose estimation has been numerous and has yielded impressive results. However, studies often focus only on improving the accuracy of the estimation results. In this paper, we propose a fast, unified end-to-end model for estimating 3D human pose, called YOLOv5-HR-TCM (YOLOv5-HRet-Temporal Convolution Model). Our proposed model is based on the 2D to 3D lifting approach for 3D human pose estimation while taking care of each step in the estimation process, such as person detection, 2D human pose estimation, and 3D human pose estimation. The proposed model is a combination of best practices at each stage. Our proposed model is evaluated on the Human 3.6M dataset and compared with other methods at each step. The method achieves high accuracy, not sacrificing processing speed. The estimated time of the whole process is 3.146 FPS on a low-end computer. In particular, we propose a sports scoring application based on the deviation angle between the estimated 3D human posture and the standard (reference) origin. The average deviation angle evaluated on the Human 3.6M dataset (Protocol #1–Pro #1) is 8.2 degrees.

show abstract

Section: Related Workmentioning

confidence: 99%

“… Illustrating of using human posture estimation in weightlifting practice ( a ) [ 5 ], healthcare, sports ( b , d ) [ 1 ], and robotics ( c ) [ 3 ]. …”

Section: Figurementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Unified End-to-End YOLOv5-HR-TCM Framework for Automatic 2D/3D Human Pose Estimation for Real-Time Applications

Nguyen¹,

Nguyen²,

Scherer

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Computer vision analysis allows the extraction of objective, quantitative measures from videos, beyond what is possible by visual assessment alone ( Davies et al, 2022 ). Pose estimation algorithms allow for real-time tracking and quantitative assessment of human movement and offer great potential for use in clinical evaluations ( Stenum et al, 2021 ).…”

Section: Introductionmentioning

confidence: 99%

Patient-led development of digital endpoints and the use of computer vision analysis in assessment of motor function in rare diseases

Ferrer-Mallol¹,

Matthews²,

Stoodley³

et al. 2022

Front. Pharmacol.

View full text Add to dashboard Cite

Digital health technologies are transforming the way health outcomes are captured and measured. Digital biomarkers may provide more objective measurements than traditional approaches as they encompass continuous and longitudinal data collection and use of automated analysis for data interpretation. In addition, the use of digital health technology allows for home-based disease assessments, which in addition to reducing patient burden from on-site hospital visits, provides a more holistic picture of how the patient feels and functions in the real world. Tools that can robustly capture drug efficacy based on disease-specific outcomes that are meaningful to patients, are going to be key to the successful development of new treatments. This is particularly important for people living with rare and chronic complex conditions, where therapeutic options are limited and need to be developed using a patient-focused approach to achieve the biggest impact. Working in partnership with patient Organisation Duchenne UK, we co-developed a video-based approach, delivered through a new mobile health platform (DMD Home), to assess motor function in patients with Duchenne muscular dystrophy (DMD), a genetic, rare, muscular disease characterized by the progressive loss of muscle function and strength. Motor function tasks were selected to reflect the “transfer stage” of the disease, when patients are no longer able to walk independently but can stand and weight-bear to transfer. This stage is important for patients and families as it represents a significant milestone in the progression of DMD but it is not routinely captured and/or scored by standard DMD clinical and physiotherapy assessments. A total of 62 videos were submitted by eight out of eleven participants who onboarded the app and were analysed with pose estimation software (OpenPose) that led to the extraction of objective, quantitative measures, including time, pattern of movement trajectory, and smoothness and symmetry of movement. Computer vision analysis of video tasks to identify voluntary or compensatory movements within the transfer stage merits further investigation. Longitudinal studies to validate DMD home as a new methodology to predict progression to the non-ambulant stage will be pursued.

show abstract

“…Systems such as exoskeletons or other devices that must be worn along the arm can be cumbersome and may lead to low patient compliance [ 11 , 12 ]. Video or depth camera arrays [ 13 , 14 ] may be objectionable for patients and home occupants due to the feeling of being under constant surveillance [ 15 ]. Traditionally, accelerometry has been used to give information about a patient’s motion in a home environment [ 16 , 17 , 18 , 19 ], but this provides only coarse measures of patient capability.…”

Section: Introductionmentioning

confidence: 99%

Capturing Upper Body Kinematics and Localization with Low-Cost Sensors for Rehabilitation Applications

Sarker

Emenonye

Kelliher

et al. 2022

Sensors

View full text Add to dashboard Cite

For upper extremity rehabilitation, quantitative measurements of a person’s capabilities during activities of daily living could provide useful information for therapists, including in telemedicine scenarios. Specifically, measurements of a person’s upper body kinematics could give information about which arm motions or movement features are in need of additional therapy, and their location within the home could give context to these motions. To that end, we present a new algorithm for identifying a person’s location in a region of interest based on a Bluetooth received signal strength (RSS) and present an experimental evaluation of this and a different Bluetooth RSS-based localization algorithm via fingerprinting. We further present algorithms for and experimental results of inferring the complete upper body kinematics based on three standalone inertial measurement unit (IMU) sensors mounted on the wrists and pelvis. Our experimental results for localization find the target location with a mean square error of 1.78 m. Our kinematics reconstruction algorithms gave lower errors with the pelvis sensor mounted on the person’s back and with individual calibrations for each test. With three standalone IMUs, the mean angular error for all of the upper body segment orientations was close to 21 degrees, and the estimated elbow and shoulder angles had mean errors of less than 4 degrees.

show abstract

Applications of Pose Estimation in Human Health and Performance across the Lifespan

Cited by 85 publications

References 126 publications

Unified End-to-End YOLOv5-HR-TCM Framework for Automatic 2D/3D Human Pose Estimation for Real-Time Applications

Unified End-to-End YOLOv5-HR-TCM Framework for Automatic 2D/3D Human Pose Estimation for Real-Time Applications

Patient-led development of digital endpoints and the use of computer vision analysis in assessment of motor function in rare diseases

Capturing Upper Body Kinematics and Localization with Low-Cost Sensors for Rehabilitation Applications

Contact Info

Product

Resources

About