Human pose estimation using deep learning: review, methodologies, progress and future research directions

Kumar, Pranjal; Chauhan, Siddhartha; Awasthi, Lalit Kumar

doi:10.1007/s13735-022-00261-6

Cited by 11 publications

(6 citation statements)

References 224 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Human pose estimation using deep learning: review, methodologies, progress and future research directions [10] The core of "intelligent robot arm based on attitude recognition" is a robot arm that can imitate human hand movements 1:1. [11] Because we use 3D reconstruction algorithms, the robot arm has the characteristics of simple manipulation, high sensitivity, high accuracy and low cost.…”

Section: Discussionmentioning

confidence: 99%

“…At present, many teams have published their results on this problem. Moulard IOP Publishing doi: 10.1088/1742-6596/2741/1/012015 2 method based on optimal motion, in which the robot can accurately imitate the motion capture and record the human motion. Huang Qiang et al proposed a similarity function for robot to imitate human motion, which is used to describe the similarity degree between robot and human motion.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Intelligent robotic arm for human pose recognition based on teleoperation system

Li,

Qiu,

Zhu

et al. 2024

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

With the rapid development of skeleton recognition, machine learning and other technologies, we will find that there are great drawbacks in the control of manipulators. Robot teleoperation refers to the inclusion of human operation in the control loop of robot control. When robots deal with complex perception and a large number of tasks, teleoperation is far superior to intelligent programming when making decisions quickly and dealing with extreme situations. The goal of this paper is to build a robotic arm teleoperation system for human motion capture, so as to solve the problems that the control accuracy of the end of the robotic arm is not high and the motion of the robotic arm is greatly affected by the difference between the human arm in the current related research, the master-slave human motion mapping algorithm is designed and extended with machine learning algorithms. We use inertial motion capture to realize teleoperation, so as to avoid the use of the terminal position and orientation control method of the hand controller to form the control command of the remote robot after tedious calculation, and it is convenient for the operator to complete the attitude tracking task in real time. The obtained attitude information has a larger range, higher sensitivity and better dynamic performance.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Intelligent robotic arm for human pose recognition based on teleoperation system

Li,

Qiu,

Zhu

et al. 2024

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

show abstract

“…Approach (2) can be seen as an extension to (1), which both seem particularly suited for unconstrained field studies. Employing algorithms for visual scene understanding (for a recent review on human pose estimation see Kumar et al, 2022 ), the manual AOI assignment in (1) is substituted by automated approaches in (2). While these two approaches dramatically economize the gaze analysis process and (2) also increase the objectivity of the GCA, two further advantages are specifically achieved by motion-capture integration, that is, (a) the possibility to add AOIs that do not appear in the currently analyzed video frame (e.g., the location where a tennis ball will be hit by the racket), and (b) a synchronized recording of the participant’s movements, which allows relating the algorithmic gaze analysis to action events (e.g., the moment of response initiation).…”

Section: Discussionmentioning

confidence: 99%

Eye-tracking technology and the dynamics of natural gaze behavior in sports: an update 2016–2022

et al. 2023

View full text Add to dashboard Cite

Updating and complementing a previous review on eye-tracking technology and the dynamics of natural gaze behavior in sports, this short review focuses on the progress concerning researched sports tasks, applied methods of gaze data collection and analysis as well as derived gaze measures for the time interval of 2016–2022. To that end, a systematic review according to the PRISMA guidelines was conducted, searching Web of Science, PubMed Central, SPORTDiscus, and ScienceDirect for the keywords: eye tracking, gaze behavio*r, eye movement, and visual search. Thirty-one studies were identified for the review. On the one hand, a generally increased research interest and a wider area of researched sports with a particular increase in official’s gaze behavior were diagnosed. On the other hand, a general lack of progress concerning sample sizes, amounts of trials, employed eye-tracking technology and gaze analysis procedures must be acknowledged. Nevertheless, first attempts to automated gaze-cue-allocations (GCA) in mobile eye-tracking studies were seen, potentially enhancing objectivity, and alleviating the burden of manual workload inherently associated with conventional gaze analyses. Reinforcing the claims of the previous review, this review concludes by describing four distinct technological approaches to automating GCA, some of which are specifically suited to tackle the validity and generalizability issues associated with the current limitations of mobile eye-tracking studies on natural gaze behavior in sports.

show abstract

“…With the advent of deep learning-based techniques that can precisely identify and categorize complicated postures, pose estimation technology has evolved tremendously recently [34] .…”

Section: Introductionmentioning

confidence: 99%

Enhancing Elderly Wellness through AI- Powered Yoga and Exercise Support Systems

Adeyemi

2024

Preprint

View full text Add to dashboard Cite

This research presents a pioneering and comprehensive approach aimed at delivering personalized exercise and yoga support to the senior population. The primary objective of this research is to enhance physical well-being, boost engagement, foster adherence to exercise and yoga routines, and refine users' posture and balance. A pivotal phase in this project involves keypoint extraction, skillfully executed using MoveNet, an advanced posture estimation model. To achieve this, deep learning algorithms such as Convolutional Neural Networks (CNNs), Dense Neural Networks (DNNs), and Multi-Layer Perceptrons (MLPs) are deployed to effectively classify and categorize yoga and exercise positions. The culmination of this endeavor manifests in the form of an interactive web application. The robustness and efficacy of the system are underscored by extensive user testing, which has also assessed its usability and potential to significantly enhance the physical well-being of the elderly. In summation, this research represents a substantial advancement in the realm of targeted exercise and yoga support, expressly designed to cater to the distinctive needs of the senior demographic.

show abstract

Human pose estimation using deep learning: review, methodologies, progress and future research directions

Cited by 11 publications

References 224 publications

Intelligent robotic arm for human pose recognition based on teleoperation system

Intelligent robotic arm for human pose recognition based on teleoperation system

Eye-tracking technology and the dynamics of natural gaze behavior in sports: an update 2016–2022

Enhancing Elderly Wellness through AI- Powered Yoga and Exercise Support Systems

Contact Info

Product

Resources

About