A Review of 3D Human Pose Estimation from 2D Images

Bartol, Kristijan; Bojanic, David C.; Petković, Tomislav; D'Apuzzo, Nicola; Pribanić, Tomislav

doi:10.15221/20.29

Cited by 9 publications

(4 citation statements)

References 59 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, HALE might lack robustness; the user must place the smartphone at a predetermined distance, height, and angle, or the app cannot correctly assess their squat. However, this issue can be resolved by using the recently developed 3D pose estimation algorithms [ 31 ]. The current version of HALE only supports squats; however, it can be extended to other exercises by collecting new data sets and training new models.…”

Section: Discussionmentioning

confidence: 99%

An Artificial Intelligence Exercise Coaching Mobile App: Development and Randomized Controlled Trial to Verify Its Effectiveness in Posture Correction

Chae¹,

Kim²,

Park³

et al. 2023

Interact J Med Res

View full text Add to dashboard Cite

Background Insufficient physical activity due to social distancing and suppressed outdoor activities increases vulnerability to diseases like cardiovascular diseases, sarcopenia, and severe COVID-19. While bodyweight exercises, such as squats, effectively boost physical activity, incorrect postures risk abnormal muscle activation joint strain, leading to ineffective sessions or even injuries. Avoiding incorrect postures is challenging for novices without expert guidance. Existing solutions for remote coaching and computer-assisted posture correction often prove costly or inefficient. Objective This study aimed to use deep neural networks to develop a personal workout assistant that offers feedback on squat postures using only mobile devices—smartphones and tablets. Deep learning mimicked experts’ visual assessments of proper exercise postures. The effectiveness of the mobile app was evaluated by comparing it with exercise videos, a popular at-home workout choice. Methods Twenty participants were recruited without squat exercise experience and divided into an experimental group (EXP) with 10 individuals aged 21.90 (SD 2.18) years and a mean BMI of 20.75 (SD 2.11) and a control group (CTL) with 10 individuals aged 22.60 (SD 1.95) years and a mean BMI of 18.72 (SD 1.23) using randomized controlled trials. A data set with over 20,000 squat videos annotated by experts was created and a deep learning model was trained using pose estimation and video classification to analyze the workout postures. Subsequently, a mobile workout assistant app, Home Alone Exercise, was developed, and a 2-week interventional study, in which the EXP used the app while the CTL only followed workout videos, showed how the app helps people improve squat exercise. Results The EXP significantly improved their squat postures evaluated by the app after 2 weeks (Pre: 0.20 vs Mid: 4.20 vs Post: 8.00, P=.001), whereas the CTL (without the app) showed no significant change in squat posture (Pre: 0.70 vs Mid: 1.30 vs Post: 3.80, P=.13). Significant differences were observed in the left (Pre: 75.06 vs Mid: 76.24 vs Post: 63.13, P=.02) and right (Pre: 71.99 vs Mid: 76.68 vs Post: 62.82, P=.03) knee joint angles in the EXP before and after exercise, with no significant effect found for the CTL in the left (Pre: 73.27 vs Mid: 74.05 vs Post: 70.70, P=.68) and right (Pre: 70.82 vs Mid: 74.02 vs Post: 70.23, P=.61) knee joint angles. Conclusions EXP participants trained with the app experienced faster improvement and learned more nuanced details of the squat exercise. The proposed mobile app, offering cost-effective self-discovery feedback, effectively taught users about squat exercises without expensive in-person trainer sessions. Trial Registration Clinical Research Information Service KCT0008178 (retrospectively registered); https://cris.nih.go.kr/cris/search/detailSearch.do/24006

show abstract

Section: Discussionmentioning

confidence: 99%

An Artificial Intelligence Exercise Coaching Mobile App: Development and Randomized Controlled Trial to Verify Its Effectiveness in Posture Correction

Chae¹,

Kim²,

Park³

et al. 2023

Interact J Med Res

View full text Add to dashboard Cite

show abstract

“…A pose can be calculated from a single or numerous frames, in a single (monocular) or multi-view (stereo) arrangement, and for a single or multiple individuals in the scene. In paper [22], an overview of traditional and deep learning-based 3D pose estimation algorithms is presented. Relevant evaluation measures, pose parametrizations, body models, and 3D human pose datasets are also highlighted.…”

Section: Photogrammetry In Biology and Medicinementioning

confidence: 99%

Low-Cost Method for 3D Body Measurement Based on Photogrammetry Using Smartphone

2022

View full text Add to dashboard Cite

This paper is focused on the possibilities of data collection via photogrammetry methods, using smartphone cameras and post-processing. The aim of this paper is to refer to progressive technologies that are part of smartphone devices, which bring more performance and variability of usage year by year. The theoretical part starts with looking to the past, describing problems of measurements and solutions invented by famous mathematicians, which we use nowadays. The following section deals with the background of measuring the human body and photogrammetry. The next section is about measuring and using calibration methods. The results section presents the architecture design of the system and a visual representation of how the application works. The result of processing a 3D person is a data object with measurements in real world metric units with minimum deviation. The conclusion is that we created our own low-cost method for 3D body measurement which partially or completely removes the shortcomings that were identified during the review of similar solutions. Our method is based on the use of open-source libraries, the use of a single smartphone mobile device and the creation of a true 3D human body model.

show abstract

“…The usage of a single RGB camera introduces several challenges to the problem of human pose estimation, such as, for example, the occurrence of occlusions and the lack of full-body images of some individuals. Furthermore, variations in clothing, body type, and camera angle can have a negative impact on the performance of the methods [4]. In Figure 1, it is possible to notice some of the challenges in human pose estimation from images captured in real and non-controlled environments (domains).…”

Section: Introductionmentioning

confidence: 99%

DUA: A Domain-Unified Approach for Cross-Dataset 3D Human Pose Estimation

Manesco,

Berretti,

Marana

2023

Sensors

View full text Add to dashboard Cite

Human pose estimation is an important Computer Vision problem, whose goal is to estimate the human body through joints. Currently, methods that employ deep learning techniques excel in the task of 2D human pose estimation. However, the use of 3D poses can bring more accurate and robust results. Since 3D pose labels can only be acquired in restricted scenarios, fully convolutional methods tend to perform poorly on the task. One strategy to solve this problem is to use 2D pose estimators, to estimate 3D poses in two steps using 2D pose inputs. Due to database acquisition constraints, the performance improvement of this strategy can only be observed in controlled environments, therefore domain adaptation techniques can be used to increase the generalization capability of the system by inserting information from synthetic domains. In this work, we propose a novel method called Domain Unified approach, aimed at solving pose misalignment problems on a cross-dataset scenario, through a combination of three modules on top of the pose estimator: pose converter, uncertainty estimator, and domain classifier. Our method led to a 44.1mm (29.24%) error reduction, when training with the SURREAL synthetic dataset and evaluating with Human3.6M over a no-adaption scenario, achieving state-of-the-art performance.

show abstract

A Review of 3D Human Pose Estimation from 2D Images

Cited by 9 publications

References 59 publications

An Artificial Intelligence Exercise Coaching Mobile App: Development and Randomized Controlled Trial to Verify Its Effectiveness in Posture Correction

An Artificial Intelligence Exercise Coaching Mobile App: Development and Randomized Controlled Trial to Verify Its Effectiveness in Posture Correction

Low-Cost Method for 3D Body Measurement Based on Photogrammetry Using Smartphone

DUA: A Domain-Unified Approach for Cross-Dataset 3D Human Pose Estimation

Contact Info

Product

Resources

About