A Survey on Deep Learning-Based 2D Human Pose Estimation Models
Source abstract
Abstract
In this article, a comprehensive survey of deep learning-based (DL-based) human pose estimation (HPE) that can help researchers in the domain of computer vision is presented. HPE is among the fastest-growing research domains of computer vision and is used in solving several problems for human endeavours. After the detailed introduction, three different human body modes followed by the main stages of HPE and two pipelines of two-dimensional (2D) HPE are presented. The details of the four components of HPE are also presented. The keypoints output format of two popular 2D HPE datasets and the most cited DL-based HPE articles from the year of breakthrough are both shown in tabular form. This study intends to highlight the limitations of published reviews and surveys respecting presenting a systematic review of the current DL-based solution to the 2D HPE model. Furthermore, a detailed and meaningful survey that will guide new and existing researchers on DL-based 2D HPE models is achieved. Finally, some future research directions in the field of HPE, such as limited data on disabled persons and multi-training DL-based models, are revealed to encourage researchers and promote the growth of HPE research.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Human Pose and Action Recognition · Context-Aware Activity Recognition Systems · Gait Recognition and Analysis
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (81)
Show source-provided references
1. Multi-view human pose estimation in human-robot interaction
2. View adaptive neural networks for high performance skeleton-based human action recognition
3. Towards single 2D image-level self-supervision for 3D human pose and shape estimation10.3390/app11209724 ↗
4. Semantic–structural graph convolutional networks for whole-body human pose estimation10.3390/info13030109 ↗
5. Efficient human pose estimation from single depth images
6. Articulated human detection with flexible mixtures of parts
7. Planetscope nanosatellites image classification using machine learning10.32604/csse.2022.023221 ↗
8. Predicting reaction based on customer’s transaction using machine learning approaches
9. CNN based automated weed detection system using UAV imagery
10. A new speech recognition model in a human-robot interaction scenario using NAO robot: Proposal and preliminary model
11. Smotednn: A novel model for air pollution forecasting and aqi classification10.32604/cmc.2022.021968 ↗
12. Deep learning-based modeling of groundwater storage change10.32604/cmc.2022.020495 ↗
13. Deep learning based supervised image classification using UAV images for forest areas classification10.1007/s12524-020-01231-3 ↗
14. Malware analysis in IoT & android systems with defensive mechanism
15. Advances in data preprocessing for bio-medical data fusion: An overview of the methods, challenges, and prospects10.1016/j.inffus.2021.07.001 ↗
16. Advances in multimodal data fusion in neuroimaging: Overview, challenges, and novel orientation
17. DeepPose: Human pose estimation via deep neural networks
18. F. Zhang, X. Zhu and C. Wang, “Single person pose estimation: A survey,” arXiv:2109.10056, 2021.
19. Multi-feature-based crowd video modeling for visual event detection10.1007/s00530-020-00652-x ↗
20. Recovering 3D human mesh from monocular images: A survey
21. X. Huang, X. Wang, W. Lv, X. Bai, X. Long et al., “PP-YOLOv2: A practical object detector,” arXiv:2104.10419, 2021.
22. 3D human pose, shape and texture from low-resolution images and videos
23. Improving object detection by label assignment distillation
24. Efficient object localization using convolutional networks
25. Head pose estimation: An extensive survey on recent techniques and applications
26. Monocular human pose estimation: A survey of deep learning-based methods10.1016/j.cviu.2019.102897 ↗
27. Deep 3D human pose estimation: A review10.1016/j.cviu.2021.103225 ↗
28. A comprehensive survey and deep learning-based approach for human recognition using ear biometric
29. Revisiting skeleton-based action recognition
30. ARHPE: Asymmetric relation-aware representation learning for head pose estimation in industrial human-computer interaction10.1109/TII.2022.3143605 ↗
31. Lite-HRNet: A lightweight high-resolution network
32. Pose-guided inflated 3D convnet for action recognition in videos
33. EfficientPose: Scalable single-person pose estimation10.1007/s10489-020-01918-7 ↗
34. TokenPose: Learning keypoint tokens for human pose estimation
35. HigherhrNet: Scale-aware representation learning for bottom-up human pose estimation
36. Distribution-aware coordinate representation for human pose estimation
37. Deep high-resolution representation learning for visual recognition
38. Deep high-resolution representation learning for human pose estimation
39. DensePose: Dense human pose estimation in the wild
40. Cascaded pyramid network for multi-person pose estimation
41. Mask R-CNN10.1109/TPAMI.2018.2844175 ↗
42. Realtime multi-person 2D pose estimation using part affinity fields
43. Stacked hourglass networks for human pose estimation
44. Convolutional pose machines
45. Flowing convnets for human pose estimation in videos
46. 2D human pose estimation: New benchmark and state of the art analysis
47. Human pose estimation from monocular images: A comprehensive survey10.3390/s16121966 ↗
48. Clustered pose and nonlinear appearance models for human pose estimation
49. Monocular 3D human pose estimation in the wild using improved CNN supervision
50. Finding human poses in videos using concurrent matching and segmentation
51. Contour people: A parameterized model of 2D articulated human shape
52. Total capture: A 3D deformation model for tracking faces, hands, and bodies
53. GHUM GHUML: Generative 3D human shape and articulated pose models
54. Poselet conditioned pictorial structures
55. Joint action recognition and pose estimation from video
56. Pictorial structures revisited: People detection and articulated pose estimation
57. Human pose estimation using body parts dependent joint regressors
58. Using k-poselets for detecting people and localizing their keypoints
59. W. Mao, Y. Ge, C. Shen, Z. Tian, X. Wang et al., “Poseur: Direct human pose regression with transformers,” arXiv:2201.07412, pp. 1–15, 2022.10.1007/978-3-031-20068-7_5 ↗
60. W. Mao, Y. Ge, C. Shen, Z. Tian, X. Wang et al., “TFPose: Direct human pose estimation with transformers,” arXiv:2103.15320, pp. 1–15, 2021.
61. PE-Former: Pose estimation transformer
62. Human pose estimation with iterative error feedback
63. A. Nibali, Z. He, S. Morgan and L. Prendergast, “Numerical coordinate regression with convolutional neural networks,” arXiv:1801.07372, pp. 1–10, 2018.
64. Multi-scale supervised network for human pose estimation
65. S. Li and X. Xiang, “Lightweight human pose estimation using heatmap-weighting loss,” arXiv:2205.10611, pp. 1–7, 2022.
66. K. Balakrishnan and D. Upadhyay, “BTranspose: Bottleneck transformers for human pose estimation with self-supervised pre-training,” arXiv:2204.10209, pp. 1–24, 2022.
67. Bottleneck transformers for visual recognition
68. I. Lifshitz, E. Fetaya and S. Ullman, “Human pose estimation using deep consensus voting,” arXiv:1603.08212, pp. 246–260, 2016.10.1007/978-3-319-46475-6_16 ↗
69. K. Zhang, R. Wu, P. Yao, K. Deng, D. Li et al., “Learning heatmap-style jigsaw puzzles provides good pretraining for 2d human pose estimation,” arXiv:2012.07101, pp. 1–13, 2020.
70. Z. Xiong, C. Wang, Y. Li, Y. Luo and Y. Cao, “Swin-pose: Swin transformer based human pose estimation,” arXiv:2201.07384, pp. 1–6, 2022.10.1109/MIPR54900.2022.00048 ↗
71. Posefix: Model-agnostic general human pose refinement network
72. DeepCut: Joint subset partition and labeling for multi person pose estimation
73. Deepercut: A deeper, stronger, and faster multi-person pose estimation model
74. Very deep convolutional networks for large-scale image recognition
75. A new feature pyramid network for object detection
76. Z. Su, M. Ye, G. Zhang, L. Dai and J. Sheng, “Improvement multi-stage model for human pose estimation,” arXiv:1902.07837, 2019.
77. Loss and loss functions for training deep learning neural networks
78. The progress of human pose estimation: A survey and taxonomy of models applied in 2D human pose estimation10.1109/ACCESS.2020.3010248 ↗
79. Microsoft COCO: Common objects in context
80. Human pose estimation using global and local normalization
81. 2D articulated human pose estimation and retrieval in (almost) unconstrained still images10.1007/s11263-012-0524-9 ↗