Action recognition based on motion of oriented magnitude patterns and feature selection

Phan, Hai-Hong; Vu, Ngoc-Son; Nguyen, Vu-Lam; Quoy, Mathias

doi:10.1049/iet-cvi.2017.0282

Cited by 16 publications

(4 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It can be applied to a variety of research cases, such as intelligent video surveillance, human-computer interaction and video content understanding. According to the different recognition targets, it can be divided into two categories: human normal behavior recognition and classification [1][2][3][4][5][6][7][8] and human abnormal behavior detection and warning [9][10][11][12][13]. With the development of research theories and methods, human normal behavior recognition and classification mainly study how to distinguish the different behaviors of human beings, and it has evolved into research on the breadth and diversity of different behaviors.…”

Section: Introductionmentioning

confidence: 99%

Human Fall Detection Based on Body Posture Spatio-Temporal Evolution

Zhang

Wang

2020

Sensors

View full text Add to dashboard Cite

Abnormal falls in public places have significant safety hazards and can easily lead to serious consequences, such as trampling by people. Vision-driven fall event detection has the huge advantage of being non-invasive. However, in actual scenes, the fall behavior is rich in diversity, resulting in strong instability in detection. Based on the study of the stability of human body dynamics, the article proposes a new model of human posture representation of fall behavior, called the “five-point inverted pendulum model”, and uses an improved two-branch multi-stage convolutional neural network (M-CNN) to extract and construct the inverted pendulum structure of human posture in real-world complex scenes. Furthermore, we consider the continuity of the fall event in time series, use multimedia analytics to observe the time series changes of human inverted pendulum structure, and construct a spatio-temporal evolution map of human posture movement. Finally, based on the integrated results of computer vision and multimedia analytics, we reveal the visual characteristics of the spatio-temporal evolution of human posture under the potentially unstable state, and explore two key features of human fall behavior: motion rotational energy and generalized force of motion. The experimental results in actual scenes show that the method has strong robustness, wide universality, and high detection accuracy.

show abstract

Section: Introductionmentioning

confidence: 99%

Human Fall Detection Based on Body Posture Spatio-Temporal Evolution

Zhang

Wang

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…It should be underlined that in all tests we process the original videos, not the stabilized ones. Using the HMDB51 dataset, the proposed approach is compared with similar approaches reported in the literature, including Phan et al [69], Zheng et al [70], and Wang et al [71]. The obtained results on the test set of HMDB51 dataset are presented in Table 2.…”

Section: Resultsmentioning

confidence: 98%

A distributed Content-Based Video Retrieval system for large datasets

Saoudi

Andaloussi

2021

J Big Data

View full text Add to dashboard Cite

With the rapid growth in the amount of video data, efficient video indexing and retrieval methods have become one of the most critical challenges in multimedia management. For this purpose, Content-Based Video Retrieval (CBVR) is nowadays an active area of research. In this article, a CBVR system providing similar videos from a large multimedia dataset based on query video has been proposed. This approach uses vector motion-based signatures to describe the visual content and uses machine learning techniques to extract key frames for rapid browsing and efficient video indexing. The proposed method has been implemented on both single machine and real-time distributed cluster to evaluate the real-time performance aspect, especially when the number and size of videos are large. Experiments were performed using various benchmark action and activity recognition datasets and the results reveal the effectiveness of the proposed method in both accuracy and processing time compared to previous studies.

show abstract

“…It is useful in a range of research scenarios, including elder surveillance, intelligent surveillance cameras, human-computer interaction, and analyzing video information [2] [3]. It can be divided into two groups based on the identification objectives: the identification and classification of normal human behavior [4] [5] and the detection and warning of abnormal human behavior [6]. Elderly people are more likely to be affected by falls that result in devastating injuries or death.…”

Section: Introductionmentioning

confidence: 99%

Human Fall Down Recognition Using Coordinates Key Points Skeleton

Ali¹,

Hussain²,

Sadiq³

2022

Int. J. Onl. Eng.

View full text Add to dashboard Cite

Falls pose a substantial threat to human safety and can quickly result in disastrous repercussions. This threat is particularly true for the elderly, where falls are the leading cause of hospitalization and injury-related death. A fall that is detected and responded to quickly has a lower danger and long-term impact. Many real-time fall detection solutions are available; however, these solutions have specific privacy, maintenance, and proper use issues. Vision-based fall event detection has the benefit of being completely private and straightforward to use and maintain. However, in real-world scenarios, falls are diverse and result in high detection instability. This study proposes a novel vision-based technique for fall detection and analyzes an extracted skeleton to define human postures. OpenPose can be used to get skeletal information about the human body. It identifies a fall using three critical parameters: the center of the value of the head and shoulder coordinates, the critical points of the shoulder coordinates, and the distance between the center of the skeleton's head and the floor with the angle between the center of the shoulders and the ground. Our proposed methodology was effective, with a classification accuracy of 97.7%.

show abstract

Action recognition based on motion of oriented magnitude patterns and feature selection

Cited by 16 publications

References 62 publications

Human Fall Detection Based on Body Posture Spatio-Temporal Evolution

Human Fall Detection Based on Body Posture Spatio-Temporal Evolution

A distributed Content-Based Video Retrieval system for large datasets

Human Fall Down Recognition Using Coordinates Key Points Skeleton

Contact Info

Product

Resources

About