Deformation-Based Abnormal Motion Detection using 3D Skeletons

Baptista, Renato; Demisse, Girum G.; Aouada, Djamila; Ottersten, Björn

doi:10.1109/ipta.2018.8608143

Cited by 7 publications

(5 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The skeletons collected using Microsoft Kinect (depth) camera has been used in the past studies [68], [69]. However, the defunct production of the Microsoft Kinect camera [70] has lead to hardware constraints in the further development of skeletal anomaly detection approaches.…”

Section: Discussionmentioning

confidence: 99%

Skeletal Video Anomaly Detection using Deep Learning: Survey, Challenges and Future Directions

Mishra¹,

Mihailidis²,

Khan³

2023

Preprint

View full text Add to dashboard Cite

The existing methods for video anomaly detection mostly utilize videos containing identifiable facial and appearance-based features. The use of videos with identifiable faces raises privacy concerns, especially when used in a hospital or community-based setting. Appearance-based features can also be sensitive to pixel-based noise, straining the anomaly detection methods to model the changes in the background and making it difficult to focus on the actions of humans in the foreground. Structural information in the form of skeletons describing the human motion in the videos is privacy-protecting and can overcome some of the problems posed by appearance-based features. In this paper, we present a survey of privacy-protecting deep learning anomaly detection methods using skeletons extracted from videos. We present a novel taxonomy of algorithms based on the various learning approaches. We conclude that skeleton-based approaches for anomaly detection can be a plausible privacyprotecting alternative for video anomaly detection. Lastly, we identify major open research questions and provide guidelines to address them.

show abstract

Section: Discussionmentioning

confidence: 99%

Skeletal Video Anomaly Detection using Deep Learning: Survey, Challenges and Future Directions

Mishra¹,

Mihailidis²,

Khan³

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The skeletons collected using Microsoft Kinect (depth) camera has been used in the past studies [87], [88]. However, the defunct production of the Microsoft Kinect camera [89] has led to hardware constraints in the further development of skeletal anomaly detection approaches.…”

Section: Hardwarementioning

confidence: 99%

Skeletal Video Anomaly Detection Using Deep Learning: Survey, Challenges, and Future Directions

Mishra,

Mihailidis,

Khan

2024

IEEE Trans. Emerg. Top. Comput. Intell.

View full text Add to dashboard Cite

The existing methods for video anomaly detection mostly utilize videos containing identifiable facial and appearancebased features. The use of videos with identifiable faces raises privacy concerns, especially when used in a hospital or communitybased setting. Appearance-based features can also be sensitive to pixel-based noise, straining the anomaly detection methods to model the changes in the background and making it difficult to focus on the actions of humans in the foreground. Structural information in the form of skeletons describing the human motion in the videos is privacy-protecting and can overcome some of the problems posed by appearance-based features. In this paper, we present a survey of privacy-protecting deep learning anomaly detection methods using skeletons extracted from videos. We present a novel taxonomy of algorithms based on the various learning approaches. We conclude that skeleton-based approaches for anomaly detection can be a plausible privacy-protecting alternative for video anomaly detection. Lastly, we identify major open research questions and provide guidelines to address them.

show abstract

“…However, while the network can easily detect surprise samples, fine-grained differences among normal samples are not discussed. Away from imagebased methods, recent studies have also been focused on systems using 3D human pose information [27], [28]. In this paper, we apply discrepancy detection both on videos and 3D human poses and discuss the ability of the system to detect fine-grained differences between two input motions.…”

Section: Discrepancy Detectionmentioning

confidence: 99%

AI Golf: Golf Swing Analysis Tool for Self-Training

2022

View full text Add to dashboard Cite

In the field of the acquisition of sports skills, a common way to improve sports skills, such as golf swing, is imitating professional players' motions. However, it is hard for beginners to specify keyframes they should focus on and which part of the body they should correct due to inconsistent timing and a lack of knowledge. In this work, a golf swing analysis tool using neural networks is proposed to address this gap. The proposed system compares two motion sequences and specifies keyframes where significant differences can be found between the two motions. Also, the system helps users intuitively understand the difference between themselves and professional players with interpretable clues. The main challenge of this work is targeting the fine-grained difference between users and professionals that can be used for self-training. Moreover, the significance of the proposed approach is using an unsupervised learning manner without prior knowledge and labeled data that will benefit future applications and research for other sports and skill training processes. In our approach, neural networks are used first to create a motion synchronizer to align motions with different phases and timing. Next, a motion discrepancy detector is implemented to find fine-grained differences between motions in latent spaces, which are learned by the networks. Furthermore, we consider that learning intermediate motion may be feasible for beginners because, in this way, they can gradually change their pose to match the ideal form. Therefore, based on the synchronization and the discrepancy detection results, we utilize a decoder to restore intermediate human poses between two motions from the latent space. Finally, we suggest possible applications for analyzing and visualizing the discrepancy between two input motions and interacting with users. With the proposed application, users can easily understand the difference between their motions and various experts' motions during self-training and learn the way they should improve their motions. INDEX TERMSComputer vision, machine learning, motor skill training I. INTRODUCTION Recent work [8] introduces a climbing training system where

show abstract

Deformation-Based Abnormal Motion Detection using 3D Skeletons

Cited by 7 publications

References 31 publications

Skeletal Video Anomaly Detection using Deep Learning: Survey, Challenges and Future Directions

Skeletal Video Anomaly Detection using Deep Learning: Survey, Challenges and Future Directions

Skeletal Video Anomaly Detection Using Deep Learning: Survey, Challenges, and Future Directions

AI Golf: Golf Swing Analysis Tool for Self-Training

Contact Info

Product

Resources

About