A Survey on Gait Recognition

Abstract

Recognizing people by their gait has become more and more popular nowadays due to the following reasons. First, gait recognition can work well remotely. Second, gait recognition can be done from low-resolution videos and with simple instrumentation. Third, gait recognition can be done without the cooperation of individuals. Fourth, gait recognition can work well while other features such as faces and fingerprints are hidden. Finally, gait features are typically difficult to be impersonated. Recent ubiquity of smartphones that capture gait patterns through accelerometers and gyroscope and advances in machine learning have opened new research directions and applications in gait recognition. A timely survey that addresses current advances is missing. In this article, we survey research works in gait recognition. In addition to recognition based on video, we address new modalities, such as recognition based on floor sensors, radars, and accelerometers; new approaches that include machine learning methods; and examine challenges and vulnerabilities in this field. In addition, we propose a set of future research directions. Our review reveals the current state-of-art and can be helpful to both experts and newcomers of gait recognition. Moreover, it lists future works and publicly available databases in gait recognition for researchers.

Evaluation of Pattern Recognition Methods for Head Gesture-Based Interface of a Virtual Reality Helmet Equipped with a Single IMU Sensor

“…In some research, authors also utilize DTW [20]. However, as can be seen in papers that contain a comprehensive survey on this subject [12,13,21,22] Principal Components Analysis (PCA) is a commonly used procedure for dimensionality reduction and features set generation for human motions analysis and classification. Because application of PCA in various steps of motion analysis and classification seems to be a standard approach for gait/full body motion analysis (although it was not mentioned in state-of-the-art papers devoted to head gestures recognition) the rest of this survey will be devoted to PCA.…”

“…Both an original data set that was used for this evaluation, Android OS software for data acquisition and all R-language source codes we used to conduct the research is available to download in order to make our research reproducible and to check methodology and implementation details (The source codes and data can be download from https://github.com/browarsoftware/headmotions). Some methods that we are comparing with our approach, although they were developed several years ago, are extensively used in up-to-date researches like for example DTW-based classification [14] or various application of PCA-based dimensionality reduction and features selection [21,32]. Also, SVM and KNN classifiers [15] are often utilized to classify data acquired by IMU sensors.…”

Calibrated and synchronized multi-view video and motion capture dataset for evaluation of gait recognition

“…Recently, deep learning methods such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs) have shown vast potential for automatically learning features for action recognition and motion analysis [4]. Several CNN-based methods have been applied to biometric gait recognition [70]. Deep learning-based methods usually benefit from big training datasets.…”

“…In the last decade, gait recognition has been extensively studied as a behavioral biometric technique [17,19,20,70]. Identity recognition on the basis of gait is non-invasive, can be done at a distance, is non-cooperative in nature, and is difficult to disguise.…”

Individuals Identification Based on Palm Vein Matching under a Parallel Environment

“…Based on the criteria, there are several distinctive characteristics, or biometric traits, which have been studied and tested. Among them are: fingerprint [4,5], face [6,7], iris [8,9], palmprint [10,11], voice [12,13], finger and palm vein [14,15], gait [16,17], signature [18], DNA [19,20], and others. An extensive review of biometric technology is presented in [21].…”