Researching | Lightweight and High-Resolution Human Pose Estimation Method

Laser & Optoelectronics Progress, Volume. 59, Issue 18, 1810012(2022)

Lightweight and High-Resolution Human Pose Estimation Method

Hanbing Qu and Zhentang Jia^*

Author Affiliations

College of Electronics and Information Engineering, Shanghai University of Electric Power, Shanghai 200090, China

show less

Abstract Get PDF(in Chinese)

Figures&Tables (13)

References (21)

Paper Information

Figures & Tables(13)

Fig. 1. Network structure

Download full size

Fig. 2. DenseNet

Download full size

Fig. 3. Improved dense network module. (a) Residual unit; (b) dense cell

Download full size

Fig. 4. Comparison of fusion methods. (a) Final fusion method of original HRNet; (b) improved final fusion method

Download full size

Fig. 5. Model training loss function curve. (a) First stage loss value; (b) second stage loss value

Download full size

Fig. 6. Comparison of prediction accuracy of various parts of different methods

Download full size

Fig. 7. Validation results. (a) Key parts overlap detection map;(b) obstacle occlusion detection map;(c) multi-person detection map

Download full size

Fig. 8. Sample analysis of insufficient model performance

Download full size

Table 1. Key points in COCO dataset

Table 1. Key points in COCO dataset

Serial number	Name	Serial number	Name
0	Nose	9	Right knee
1	Neck	10	Right ankle
2	Right shoulder	11	Left hip
3	Right elbow	12	Left knee
4	Right wrist	13	Left ankle
5	Left shoulder	14	Right eye
6	Left elbow	15	Left eye
7	Left wrist	16	Right ear
8	Right crotch	17	Left ear

Table 2. Validation and comparison of different methods under COCO dataset

Table 2. Validation and comparison of different methods under COCO dataset

Method	#Params /MB	GFLOPs	Model size /MB	AP /%	AP⁵⁰ /%	AP⁷⁵ /%	AP^L /%	AP^M /%
CPN	27.0	6.2	314.0	68.6
SimpleBaseline‐50	34.0	9.0	129.0	70.4	88.6	78.3	67.1	77.2
SimpleBaseline‐101	53.0	12.4	202.0	71.4	89.3	79.3	68.1	78.1
HRNetV1	28.5	16.0	109.4	74.9	92.5	82.8	71.3	80.9
HigherHRNet	28.6	47.9	109.8	66.4	87.5	72.8	61.2	74.2
HRNet	28.5	7.1	109.0	74.4	90.5	81.9	70.8	81.0
Proposed method	10.1	6.5	30.6	74.8	92.6	83.2	72.2	77.7

Table 3. Comparison of different methods under MPII dataset

Table 3. Comparison of different methods under MPII dataset

Method	Head	Shoulder	Elbow	Wrist	Crotch	Lap	Ankle	Mean
DeeperCut^［21］	97.2	94.5	87.3	82.4	86.2	81.7	77.2	86.6
SimpleBaseline‐50	96.4	95.3	89.0	83.2	88.4	84.0	79.6	88.5
SimpleBaseline‐101	96.9	95.9	89.5	84.4	88.4	84.5	80.7	89.1
HRNet	97.1	95.9	90.3	86.4	89.1	87.1	83.3	90.3
Proposed method	97.8	95.4	90.1	83.9	88.5	87.2	82.8	88.9

Table 4. Ablation experiment under COCO validation dataset
View table
Table 4. Ablation experiment under COCO validation dataset
Model #Params /MB AP /%
HRNet 28.5 74.4
Proposed（A） 9.8 73.6
Proposed（B） 10.1 74.8

Table 5. Speed real-time comparison
View table
Table 5. Speed real-time comparison
Model Training time /h Single image detection time /ms Accuracy （PCK） /%
CPMs 131 106.3 85.5
HRNet 54 52.2 90.3
Proposed model 45 24.0 88.9

Tools

Get Citation

Copy Citation Text

Hanbing Qu, Zhentang Jia. Lightweight and High-Resolution Human Pose Estimation Method[J]. Laser & Optoelectronics Progress, 2022, 59(18): 1810012

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Image Processing

Received: May. 27, 2021

Accepted: Aug. 10, 2021

Published Online: Aug. 22, 2022

The Author Email: Jia Zhentang (462458081@qq.com)

DOI:10.3788/LOP202259.1810012

Topics

laser devices and laser physics

Lasers and Laser Optics

laser manufacturing

Instrumentation, Measurement and Metrology