Robust Gait Recognition by Integrating Inertial and RGBD Sensors

Zou, Qin; Ni, Lihao; Wang, Qian; Li, Qingquan; Wang, Song

doi:10.1109/tcyb.2017.2682280

Cited by 96 publications

(42 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A one-vs-all SVM classifier is employed. • EigenGait [7]: the inertial gait data are decomposed in the eigen space, and the principle components are taken as gait features for person identification [7]. A one-vs-all SVM classifier is employed.…”

Section: E User-identification Performancementioning

confidence: 99%

Deep Learning-Based Gait Recognition Using Smartphones in the Wild

Zou

Wang

et al. 2020

IEEE Trans.Inform.Forensic Secur.

Self Cite

220

141

View full text Add to dashboard Cite

Comparing with other biometrics, gait has advantages of being unobtrusive and difficult to conceal. Inertial sensors such as accelerometer and gyroscope are often used to capture gait dynamics. Nowadays, these inertial sensors have commonly been integrated in smartphones and widely used by average person, which makes it very convenient and inexpensive to collect gait data. In this paper, we study gait recognition using smartphones in the wild. Unlike traditional methods that often require the person to walk along a specified road and/or at a normal walking speed, the proposed method collects inertial gait data under a condition of unconstraint without knowing when, where, and how the user walks. To obtain a high performance of person identification and authentication, deep-learning techniques are presented to learn and model the gait biometrics from the walking data. Specifically, a hybrid deep neural network is proposed for robust gait feature representation, where features in the space domain and in the time domain are successively abstracted by a convolutional neural network and a recurrent neural network. In the experiments, two datasets collected by smartphones on a total of 118 subjects are used for evaluations. Experiments show that the proposed method achieves over 93.5% and 93.7% accuracy in person identification and authentication, respectively.Index Terms-Gait recognition, inertial sensor, person identification, convolutional neural network, recurrent neural network.

show abstract

Section: E User-identification Performancementioning

confidence: 99%

Deep Learning-Based Gait Recognition Using Smartphones in the Wild

Zou

Wang

et al. 2020

IEEE Trans.Inform.Forensic Secur.

Self Cite

220

141

View full text Add to dashboard Cite

show abstract

“…The sensors used for motion analysis, such as optical sensors, inertial sensor units, force platforms and balance boards, are less intrusive, easy to deploy in the experimental work and efficiently acquire data. It is evident that fusion of data obtained from various sources (for example from inertial and Kinect systems) has produced better results in identifying data patterns [77,78]. To integrate multiple sensor data at feature level or to fuse those at decision level, may increase the overall accuracy of a general classifier for identifying various frailty levels.…”

Section: Future Directionsmentioning

confidence: 99%

Assessment of frailty: a survey of quantitative and clinical methods

Panhwar

Naghdy

et al. 2019

BMC biomed eng

View full text Add to dashboard Cite

Background: Frailty assessment is a critical approach in assessing the health status of older people. The clinical tools deployed by geriatricians to assess frailty can be grouped into two categories; using a questionnaire-based method or analyzing the physical performance of the subject. In performance analysis, the time taken by a subject to complete a physical task such as walking over a specific distance, typically three meters, is measured. The questionnaire-based method is subjective, and the time-based performance analysis does not necessarily identify the kinematic characteristics of motion and their root causes. However, kinematic characteristics are crucial in measuring the degree of frailty. Results: The studies reviewed in this paper indicate that the quantitative analysis of activity of daily living, balance and gait are significant methods for assessing frailty in older people. Kinematic parameters (such as gait speed) and sensor-derived parameters are also strong markers of frailty. Seventeen gait parameters are found to be sensitive for discriminating various frailty levels. Gait velocity is the most significant parameter. Short term monitoring of daily activities is a more significant method for frailty assessment than is long term monitoring and can be implemented easily using clinical tests such as sit to stand or stand to sit. The risk of fall can be considered an outcome of frailty. Conclusion: Frailty is a multi-dimensional phenomenon that is defined by various domains; physical, social, psychological and environmental. The physical domain has proven to be essential in the objective determination of the degree of frailty in older people. The deployment of inertial sensor in clinical tests is an effective method for the objective assessment of frailty.

show abstract

“…In the last few years, approaches built on deep neural networks have achieved state-of-the-art performance on many vision tasks [15]- [17] as comparing to traditional methods [45]. Inspired by the powerful representation ability of deep neural networks, some deep hashing methods have been proposed, which show great progress compared with traditional hand-crafted feature based methods.…”

Section: Related Workmentioning

confidence: 99%

Improved Deep Hashing With Soft Pairwise Similarity for Multi-Label Image Retrieval

Zheng

Zou

Lin

et al. 2020

IEEE Trans. Multimedia

Self Cite

132

View full text Add to dashboard Cite

Hash coding has been widely used in the approximate nearest neighbor search for large-scale image retrieval. Recently, many deep hashing methods have been proposed and shown largely improved performance over traditional featurelearning methods. Most of these methods examine the pairwise similarity on the semantic-level labels, where the pairwise similarity is generally defined in a hard-assignment way. That is, the pairwise similarity is '1' if they share no less than one class label and '0' if they do not share any. However, such similarity definition cannot reflect the similarity ranking for pairwise images that hold multiple labels. In this paper, an improved deep hashing method is proposed to enhance the ability of multi-label image retrieval. We introduce a pairwise quantified similarity calculated on the normalized semantic labels. Based on this, we divide the pairwise similarity into two situations -'hard similarity' and 'soft similarity', where cross-entropy loss and mean square error loss are adapted respectively for more robust feature learning and hash coding. Experiments on four popular datasets demonstrate that, the proposed method outperforms the competing methods and achieves the state-of-the-art performance in multi-label image retrieval.Index Terms-image retrieval, convolutional neural network, semantic label, pairwise similarity, deep hashing.

show abstract

Robust Gait Recognition by Integrating Inertial and RGBD Sensors

Cited by 96 publications

References 68 publications

Deep Learning-Based Gait Recognition Using Smartphones in the Wild

Deep Learning-Based Gait Recognition Using Smartphones in the Wild

Assessment of frailty: a survey of quantitative and clinical methods

Improved Deep Hashing With Soft Pairwise Similarity for Multi-Label Image Retrieval

Contact Info

Product

Resources

About