Classification of gait anomalies from kinect

Li, Qiannan; Wang, Yafang; Sharf, Andrei; Cao, Ya Bin; Tu, Changhe; Chen, Baoquan; Yu, Shengyuan

doi:10.1007/s00371-016-1330-0

Cited by 55 publications

(47 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many studies on skeleton-based abnormal gait recognition have been published [12]- [21]. However, there are a few publicly accessible 3D skeleton datasets of abnormal gaits [12], [17], [18], [46].…”

Section: A 3d Skeleton Datasets Of Abnormal Gaitsmentioning

confidence: 99%

“…human identification [7]- [11], and abnormal gait recognition [12]- [21] have been proposed in the past few years.…”

Section: Introductionmentioning

confidence: 99%

“…Additionally, skeleton-based abnormal gait recognition has also received much attention. Many methods using machine learning algorithms to recognize normal and abnormal gait patterns have been developed [12]- [21].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Feature Extraction Using an RNN Autoencoder for Skeleton-Based Abnormal Gait Recognition

Jun

Lee

et al. 2020

IEEE Access

View full text Add to dashboard Cite

In skeleton-based abnormal gait recognition, using original skeleton data decreases the recognition performance because they contain noise and irrelevant information. Instead of feeding original skeletal gait data to a recognition model, features extracted from the skeleton data are normally used. However, existing feature extraction methods might include laborious processes and it is hard for them to minimize the irrelevant information while preserving the important information. To solve this problem, an automatic feature extraction method using a recurrent neural network (RNN)-based Autoencoder (AE) is proposed in this paper. We extracted features from skeletal gait data by using two RNN AEs: a long short-term memory (LSTM)-based AE (LSTM AE) and a gated recurrent unit (GRU)-based AE (GRU AE). The features of the RNN AEs are compared to the original skeleton data and other existing features. We evaluated the features by feeding them to various discriminative models (DMs) and comparing the recognition performances. The features extracted by using the RNN AEs are more easily recognized and robust than the original skeleton data and other existing features. In particular, the LSTM AE shows a better performance than the GRU AE. Compared to single DMs fed with the original skeleton directly, hybrid models where the features of the RNN AEs are fed to DMs show a higher recognition accuracy with fewer training epochs and learning parameters. Therefore, the proposed automatic feature extraction method improves the performance of skeleton-based abnormal gait recognition by reducing laborious processes and increasing the recognition accuracy effectively.

show abstract

Section: A 3d Skeleton Datasets Of Abnormal Gaitsmentioning

confidence: 99%

“…human identification [7]- [11], and abnormal gait recognition [12]- [21] have been proposed in the past few years.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Feature Extraction Using an RNN Autoencoder for Skeleton-Based Abnormal Gait Recognition

Jun

Lee

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Many papers recently deal with gait recognition [24] and neural diseases detection [25] using a Kinect (2G). This approach eliminates common disadvantages but commonly deals with a quite short range of the Kinect sensor.…”

Section: Gait Monitoringmentioning

confidence: 99%

Robot-Based Image Analysis for Evaluating Rehabilitation after Brain Surgery

Kohout¹,

Crha²,

Trnková³

et al. 2018

mendel

View full text Add to dashboard Cite

After certain types of brain surgery, patients are often affected by changes in both their dynamic balance and facial disorder. Because rehabilitation takes several months, it is important that both doctors and patients are able to monitor progress quantitatively. At present, such quantification is subjective and highly dependent on the doctor’s opinion. Thus, we here investigate the use of robot-based image analysis for measuring rehabilitation. To evaluate a patient’s dynamic balance, we developed a mobile robotic platform that uses a stereovision camera (MS Kinect) to capture a video of the subject walking along a hospital corridor. To evaluate a patient’s facial disorders, the same camera is used in a static mode to detect and capture precise facial movements that the subject is asked to perform. From these videos, specific patterns can be extracted for rehabilitation process description.

show abstract

“…Based on the covariance matrices provided by our research partners, the Kinect system has the potential to be a low-cost home-based tool that can be used for recognition purposes in stroke and PD patients as well as healthy individuals. 4 Kinect has a relatively accurate measurement as shown by formal studies had showed 3 ; the present study therefore attempted to identify a simpler and easily available parameter to use to perform clinical assessments. Normalized SL and GV have been used to distinguish among PD patients, healthy adults, and young students, 1 so these measures represent accessible and measureable parameters.…”

Section: Kinect-based Gait Analyses Of Patients With Parkinson's Disementioning

confidence: 99%

Classification of gait anomalies from kinect

Cited by 55 publications

References 33 publications

Feature Extraction Using an RNN Autoencoder for Skeleton-Based Abnormal Gait Recognition

Feature Extraction Using an RNN Autoencoder for Skeleton-Based Abnormal Gait Recognition

Robot-Based Image Analysis for Evaluating Rehabilitation after Brain Surgery

Kinect‐based gait analyses of patients with Parkinson's disease, patients with stroke with hemiplegia, and healthy adults

Contact Info

Product

Resources

About