Person identification from partial gait cycle using fully convolutional neural networks

Babaee, Mohammadreza; Li, Linwei; Rigoll, Gerhard

doi:10.1016/j.neucom.2019.01.091

Cited by 44 publications

(13 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is well-founded to identify different persons by gait, because each person exhibits his/her walking pattern in a repeatable and sufficiently unique manner (Winter, 2009). The natural walking of a person is periodic (Babaee, Li & Rigoll, 2019). However, in a full gait cycle, the walking pattern of a person will gradually change as time goes by.…”

Section: Proposed Methodsmentioning

confidence: 99%

“…Based on the feature template, a more effective gait representation can be reconstructed using some reconstruction techniques. For example, by using ITCNet proposed in Babaee, Li & Rigoll (2019), a complete GEI of a full gait cycle will be progressively reconstructed from an incomplete GEI of a low frame-rate gait sequence. However, as we stated above, this method merely focuses on optimizing the reconstruction performance of GEI, thus its recognition accuracy cannot be guaranteed.…”

Section: Related Work Low Frame-rate Gait Recognitionmentioning

confidence: 99%

“…In Guan, Li & Choudhury (2013), gait features are directly formulated from the input low frame-rate sequences via metric learning methods. In Babaee, Li & Rigoll (2019), a CNN-based network is proposed to progressively reconstruct a complete GEI of a full gait cycle from an incomplete GEI of a low frame-rate gait sequence. In this paper, however, we give more attention to the problem of insufficient input data.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Gait recognition using a few gait frames

Yao

Kusakunniran

et al. 2021

PeerJ Computer Science

View full text Add to dashboard Cite

Gait has been deemed as an alternative biometric in video-based surveillance applications, since it can be used to recognize individuals from a far distance without their interaction and cooperation. Recently, many gait recognition methods have been proposed, aiming at reducing the influence caused by exterior factors. However, most of these methods are developed based on sufficient input gait frames, and their recognition performance will sharply decrease if the frame number drops. In the real-world scenario, it is impossible to always obtain a sufficient number of gait frames for each subject due to many reasons, e.g., occlusion and illumination. Therefore, it is necessary to improve the gait recognition performance when the available gait frames are limited. This paper starts with three different strategies, aiming at producing more input frames and eliminating the generalization error cause by insufficient input data. Meanwhile, a two-branch network is also proposed in this paper to formulate robust gait representations from the original and new generated input gait frames. According to our experiments, under the limited gait frames being used, it was verified that the proposed method can achieve a reliable performance for gait recognition.

show abstract

Section: Proposed Methodsmentioning

confidence: 99%

Section: Related Work Low Frame-rate Gait Recognitionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Gait recognition using a few gait frames

Yao

Kusakunniran

et al. 2021

PeerJ Computer Science

View full text Add to dashboard Cite

show abstract

“…The model of human silhouette release and gait recognition with a single frame, and trained them in an end-to-end manner. Maryam Babaee [14] had used a fully convolutional neural network (deep learning) for identifying the human being by partial gait cycle. A method is tested on OULP and Casia-B and gives an accuracy of full Gait Energy Image (GEI) is 96.1%.…”

Section: Literature Reviewmentioning

confidence: 99%

A Survey on Learning-Based Gait Recognition for Human Authentication in Smart Cities

Singh

Dwivedi

2021

Intelligent Sustainable Systems

View full text Add to dashboard Cite

Smart cities include good infrastructure, better transportation, connectivity, security, etc. Human identification is a part of providing security to societies. Artificial intelligence plays an important role in this objective. In this paper, we have presented a state-of-the-art on different machine learning techniques used for the identification of human beings according to their movement. We found that there are various machine learning techniques, viz., support vector machine (SVM), k-nearest neighbor (k-NN), convolutional neural network (CNN), Fuzzy set theory, Discrete Fourier transform which are used for human authentication. We presented a comparative study of such schemes and provided our major findings of the survey. We observed that under certain conditions, neural network-based techniques are performing better than the other existing schemes of gait-based human authentication.

show abstract

“…To date, gait analysis has been utilised extensively for security applications, for example, person detection and tracking [8] or as a form of person identification [9] [10]. Implications of this suggest that gait representations are unique to individual people and can be used to extract identifiable biometric information from someone's recorded gait.…”

Section: Introductionmentioning

confidence: 99%

Effects of Intra-Subject Variation in Gait Analysis on ASD Classification Performance in Machine Learning Models

Henderson

Yogarajah

Gardiner

et al. 2020

2020 31st Irish Signals and Systems Conference (ISSC)

View full text Add to dashboard Cite

Autism Spectrum Disorder (ASD) is a developmental disorder that is prevalent globally. Research into detecting autism traditionally focused on behavioural aspects of the condition, however, more recently, focus has shifted to more objective alternatives using techniques such as machine learning and gait analysis. Gait measurements, having been used for person identification, varies from person to person, introducing a lot of intra-subject variance. This applies to the 8 spatialtemporal features used in this study, representing the time that an individual spends in each phase of a gait cycle, collected using a Vicon motion tracking system. The features were averaged across each gait trial that the subjects performed, producing a second set of features with reduced intra-subject variance. Four common classifiers, a Support Vector Machine (SVM), K-Nearest Neighbour (KNN), Random Forests (RF) and a Decision Tree (DT) classifier, were all trained using the two feature sets and their classification rates were compared. The results show that for the RF classifier, reducing the intra-subject variance, was able to successfully increase the classification power. The KNN and DT classifiers experienced a minimal decrease in accuracy, where the SVM suffered the greatest loss when intrasubject variance was reduced. Results overall show that the effect intra-subject variance has on classification power depends heavily on the suitability of the classifier to the initial problem as well as size and class balance of the data.

show abstract

Person identification from partial gait cycle using fully convolutional neural networks

Cited by 44 publications

References 35 publications

Gait recognition using a few gait frames

Gait recognition using a few gait frames

A Survey on Learning-Based Gait Recognition for Human Authentication in Smart Cities

Effects of Intra-Subject Variation in Gait Analysis on ASD Classification Performance in Machine Learning Models

Contact Info

Product

Resources

About