A Survey of Human Action Recognition and Posture Prediction

Ma, Nan; Wu, Zhixuan; Cheung, Yiu-ming; Guo, Yuchen; Gao, Yue; Li, Jiahong; Jiang, Beijyan

doi:10.26599/tst.2021.9010068

Cited by 31 publications

(12 citation statements)

References 262 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Though FCNs suffer from vanishing gradient problem, this problem can be eliminated by using residual connections in the network. These models fail to e ciently learn temporal information as already proved in various studies [19].…”

Section: Methodsmentioning

confidence: 91%

“…Firstly, improving the accuracy and robustness of 3D pose estimation [18] algorithms remains a crucial direction. While deep learning-based models have shown promising results, there is still room for enhancing their performance, especially in challenging scenarios such as occlusions, complex poses [19], and cluttered backgrounds. Additionally, exploring novel architectures and fusion strategies that can effectively leverage both spatial and temporal information could lead to further improvements [6].…”

Section: Discussionmentioning

confidence: 99%

“…Each joint can be considered as a node which represents the current trajectory of that particular joint. As already proven, raw text or data is not rich in features and hence we require a feature extractor module [19]. The input x ∈ R L×17×2 consists of sequence of frames, each frame has 17 (x, y) coordinates for each joint of the body.…”

Section: Model Architecturementioning

confidence: 99%

See 2 more Smart Citations

Graph-based Semi-supervised Learning for 3D Pose Estimation

Sharma,

Jha,

Meena

2024

Preprint

View full text Add to dashboard Cite

The challenge of estimating 3D pose has been thoroughly explored and researched in computer vision due to its broad range of applications. However, due to complex structures, occlusion, frame rates, varying sizes and resolutions, this problem is highly challenging. This paper demonstrates the ability of graph neural networks (GNNs) and long short-term memory (LSTM) for 2D to 3D pose estimation in a sequence of frames. LSTM is used as a feature extractor. The bone length and joint angle play a crucial role in pose estimation. Our model uses a GNN to analyze the relationships between nearby joints and their angles and then predicts the final 3D joint positions using a multi-layer perceptron (MLP) model. A Semi-supervised learning and frame-dropping strategy approach is employed to enhance the accuracy of our model. Our model outperforms various latest advancements models, achieving an accuracy of 6.5 mm in joint localization on the HumanEva-1 and Human3.6m datasets.

show abstract

Section: Methodsmentioning

confidence: 91%

Section: Discussionmentioning

confidence: 99%

Section: Model Architecturementioning

confidence: 99%

See 1 more Smart Citation

Graph-based Semi-supervised Learning for 3D Pose Estimation

Sharma,

Jha,

Meena

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…However, overcoming factors such as occlusion, inadequate training data, and depth blur is still difficult. The commonly used posture recognition algorithms [ 143 , 152 ] in recent years are shown in Table 5 .…”

Section: Analysis and Discussionmentioning

confidence: 99%

A Survey on Artificial Intelligence in Posture Recognition

Jiang¹,

Hu²,

Wang³

et al. 2023

Computer Modeling in Engineering &Amp; Sciences

View full text Add to dashboard Cite

Over the years, the continuous development of new technology has promoted research in the field of posture recognition and also made the application field of posture recognition have been greatly expanded. The purpose of this paper is to introduce the latest methods of posture recognition and review the various techniques and algorithms of posture recognition in recent years, such as scale-invariant feature transform, histogram of oriented gradients, support vector machine (SVM), Gaussian mixture model, dynamic time warping, hidden Markov model (HMM), lightweight network, convolutional neural network (CNN). We also investigate improved methods of CNN, such as stacked hourglass networks, multi-stage pose estimation networks, convolutional pose machines, and high-resolution nets. The general process and datasets of posture recognition are analyzed and summarized, and several improved CNN methods and three main recognition techniques are compared. In addition, the applications of advanced neural networks in posture recognition, such as transfer learning, ensemble learning, graph neural networks, and explainable deep neural networks, are introduced. It was found that CNN has achieved great success in posture recognition and is favored by researchers. Still, a more in-depth research is needed in feature extraction, information fusion, and other aspects. Among classification methods, HMM and SVM are the most widely used, and lightweight network gradually attracts the attention of researchers. In addition, due to the lack of 3D benchmark data sets, data generation is a critical research direction.

show abstract

“…Challenges and trends in HAR and posture prediction are discussed by Ma et al [36] discuss, highlighting four main challenges: significant intra-class variation and inter-class similarity, complex scenarios, long untrimmed sequences, and long-tailed distributions in data. They review various datasets, methods, and algorithms and discuss recent advancements and future research directions.…”

Section: Review Of Related Work and Rationale For This Comprehensive ...mentioning

confidence: 99%

Human Action Recognition in Smart Living Services and Applications: Context Awareness, Data Availability, Personalization, and Privacy

Diraco¹,

Rescio²,

Caroppo³

et al. 2023

Preprint

View full text Add to dashboard Cite

Smart Living, an increasingly prominent concept, entails incorporating sophisticated technologies in homes and urban environments to elevate the quality of life for citizens. A critical success factor for Smart Living services and applications, from energy management to healthcare and transportation, is the efficacy of human action recognition (HAR). HAR, rooted in computer vision, seeks to identify human actions and activities using visual data and various sensor modalities. This paper extensively reviews the literature on HAR in Smart Living services and applications, amalgamating key contributions and challenges while providing insights into future research directions. The review delves into the essential aspects of Smart Living, the state of the art in HAR, and the potential societal implications of this technology. Moreover, the paper meticulously examines the primary application sectors in Smart Living that stand to gain from HAR, such as smart homes, smart healthcare, and smart cities. By underscoring the significance of the four dimensions of Context Awareness, Data Availability, Personalization, and Privacy in HAR, this paper serves as a valuable resource for researchers and practitioners striving to advance Smart Living services and applications.

show abstract

A Survey of Human Action Recognition and Posture Prediction

Cited by 31 publications

References 262 publications

Graph-based Semi-supervised Learning for 3D Pose Estimation

Graph-based Semi-supervised Learning for 3D Pose Estimation

A Survey on Artificial Intelligence in Posture Recognition

Human Action Recognition in Smart Living Services and Applications: Context Awareness, Data Availability, Personalization, and Privacy

Contact Info

Product

Resources

About