HP-YOLO: A Lightweight Real-Time Human Pose Estimation Method

Human Pose Estimation (HPE) plays a critical role in medical applications, particularly within nursing robotics for patient monitoring. Despite its importance, HPE faces several challenges, including high rates of false positives and negatives, stringent real-time requirements, and limited computati...

Full description

Saved in:
Bibliographic Details
Main Authors: Haiyan Tu, Zhengkun Qiu, Kang Yang, Xiaoyue Tan, Xiujuan Zheng
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/15/6/3025
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850203791939338240
author Haiyan Tu
Zhengkun Qiu
Kang Yang
Xiaoyue Tan
Xiujuan Zheng
author_facet Haiyan Tu
Zhengkun Qiu
Kang Yang
Xiaoyue Tan
Xiujuan Zheng
author_sort Haiyan Tu
collection DOAJ
description Human Pose Estimation (HPE) plays a critical role in medical applications, particularly within nursing robotics for patient monitoring. Despite its importance, HPE faces several challenges, including high rates of false positives and negatives, stringent real-time requirements, and limited computational resources, especially in complex backgrounds. In response, we introduce the HP-YOLO model, developed using the YOLOv8 framework, to effectively address these issues. We designed an Enhanced Large Separated Kernel Attention (ELSKA) mechanism and integrated it into the backbone network, thereby improving the model’s effective receptive field and feature separation capabilities, which enhances keypoint detection accuracy in challenging environments. Additionally, the Reparameterized Network with Cross-Stage Partial Connections and Efficient Layer Aggregation Network (RepNCSPELAN4) module was incorporated into the detection head, boosting accuracy in detecting small-sized targets through multi-scale convolution and reparameterization techniques while accelerating inference speed. On the COCO dataset, our HP-YOLO model outperformed existing lightweight methods by increasing average precision (AP) by 4.9%, while using 18% fewer parameters and achieving 1.4× higher inference speed. Our method significantly enhances the real-time performance and efficiency of human pose estimation while maintaining high accuracy, offering an optimal solution for applications in complex environments.
format Article
id doaj-art-2d15d567e9e642008c88fe01f4854e24
institution OA Journals
issn 2076-3417
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Applied Sciences
spelling doaj-art-2d15d567e9e642008c88fe01f4854e242025-08-20T02:11:25ZengMDPI AGApplied Sciences2076-34172025-03-01156302510.3390/app15063025HP-YOLO: A Lightweight Real-Time Human Pose Estimation MethodHaiyan Tu0Zhengkun Qiu1Kang Yang2Xiaoyue Tan3Xiujuan Zheng4Department of Automation, College of Electrical Engineering, Sichuan University, Chengdu 610065, ChinaDepartment of Automation, College of Electrical Engineering, Sichuan University, Chengdu 610065, ChinaDepartment of Automation, College of Electrical Engineering, Sichuan University, Chengdu 610065, ChinaDepartment of Automation, College of Electrical Engineering, Sichuan University, Chengdu 610065, ChinaDepartment of Automation, College of Electrical Engineering, Sichuan University, Chengdu 610065, ChinaHuman Pose Estimation (HPE) plays a critical role in medical applications, particularly within nursing robotics for patient monitoring. Despite its importance, HPE faces several challenges, including high rates of false positives and negatives, stringent real-time requirements, and limited computational resources, especially in complex backgrounds. In response, we introduce the HP-YOLO model, developed using the YOLOv8 framework, to effectively address these issues. We designed an Enhanced Large Separated Kernel Attention (ELSKA) mechanism and integrated it into the backbone network, thereby improving the model’s effective receptive field and feature separation capabilities, which enhances keypoint detection accuracy in challenging environments. Additionally, the Reparameterized Network with Cross-Stage Partial Connections and Efficient Layer Aggregation Network (RepNCSPELAN4) module was incorporated into the detection head, boosting accuracy in detecting small-sized targets through multi-scale convolution and reparameterization techniques while accelerating inference speed. On the COCO dataset, our HP-YOLO model outperformed existing lightweight methods by increasing average precision (AP) by 4.9%, while using 18% fewer parameters and achieving 1.4× higher inference speed. Our method significantly enhances the real-time performance and efficiency of human pose estimation while maintaining high accuracy, offering an optimal solution for applications in complex environments.https://www.mdpi.com/2076-3417/15/6/3025HPEattention mechanismmodel pruninglightweight structure
spellingShingle Haiyan Tu
Zhengkun Qiu
Kang Yang
Xiaoyue Tan
Xiujuan Zheng
HP-YOLO: A Lightweight Real-Time Human Pose Estimation Method
Applied Sciences
HPE
attention mechanism
model pruning
lightweight structure
title HP-YOLO: A Lightweight Real-Time Human Pose Estimation Method
title_full HP-YOLO: A Lightweight Real-Time Human Pose Estimation Method
title_fullStr HP-YOLO: A Lightweight Real-Time Human Pose Estimation Method
title_full_unstemmed HP-YOLO: A Lightweight Real-Time Human Pose Estimation Method
title_short HP-YOLO: A Lightweight Real-Time Human Pose Estimation Method
title_sort hp yolo a lightweight real time human pose estimation method
topic HPE
attention mechanism
model pruning
lightweight structure
url https://www.mdpi.com/2076-3417/15/6/3025
work_keys_str_mv AT haiyantu hpyoloalightweightrealtimehumanposeestimationmethod
AT zhengkunqiu hpyoloalightweightrealtimehumanposeestimationmethod
AT kangyang hpyoloalightweightrealtimehumanposeestimationmethod
AT xiaoyuetan hpyoloalightweightrealtimehumanposeestimationmethod
AT xiujuanzheng hpyoloalightweightrealtimehumanposeestimationmethod