Gait Recognition via Disentangled Representation Learning

Abstract

Gait, the walking pattern of individuals, is one of the most important biometrics modalities. Most of the existing gait recognition methods take silhouettes or articulated body models as the gait features. These methods suffer from degraded recognition performance when handling confounding variables, such as clothing, carrying and view angle. To remedy this issue, we propose a novel AutoEncoder framework to explicitly disentangle pose and appearance features from RGB imagery and the LSTM-based integration of pose features over time produces the gait feature. In addition, we collect a Frontal-View Gait (FVG) dataset to focus on gait recognition from frontal-view walking, which is a challenging problem since it contains minimal gait cues compared to other views. FVG also includes other important variations, e.g., walking speed, carrying, and clothing. With extensive experiments on CASIA-B, USF and FVG datasets, our method demonstrates superior performance to the state of the arts quantitatively, the ability of feature disentanglement qualitatively, and promising computational efficiency.

Can Ensemble Deep Learning Identify People by Their Gait Using Data Collected from Multi-Modal Sensors in Their Insole?

“…Gait analysis as a means of identifying individuals, i.e., biometrics, began by utilizing vision sensors [ 32 ]; these approaches have since been further studied and evolved [ 33 , 34 , 35 , 36 ]. However, vision-based gait analysis requires strict conditions during sensing, for example the video sequences must include only the individual to be identified, and its identification accuracy is not high enough to use it as biometric tool by itself.…”

Joint Metric Learning and Hierarchical Network for Gait Recognition

“…Most of the current methods for gait recognition enhance the discrimination of visual features. Zhang et al [7] presented a method based on automatic encoding by separates appearance and posture features from RGB frames, and then combines them together to form a gait feature that is a video segment using LSTM. To extract discriminative gait features, Yu et al [11] presented a GaitGAN method based on generation of antagonistic networks.…”

“…These methods can be performed easily, while losing temporal and spatial information. And extracts the gait features [6][7][8] directly from the original gait silhouette sequences, such as using 3D-CNN [9,10], which is a better way to learn the features. Previous methods [6,7]have been used loss function to improve discriminative feature, while these methods use only one loss function, the classification effect is not obvious, to illustrate this more explicitly, compared to using triplet or cross-entropy loss alone, the combination of two loss is more efficient, as shown in Fig.…”

User Identification from Gait Analysis Using Multi-Modal Sensors in Smart Insole

“…Gait pattern analysis comprises a sensor module for acquiring data and an application module for analyzing the data [8]. Different types of sensors are utilized in gait analysis, for instance video recorders [9], electromyography sensors [10], pressure sensors [11], accelerometers [12,13], and gyroscopes [14,15]. Initially, the gait pattern analyses were conducted in restricted environments because of the size of sensors, the inconvenience of installing sensors, and other limitations.…”

“…As gait patterns exhibit specific characteristics according to the individual, they can also be used for user identification if used along with other biometric techniques, such as face or fingerprint recognition. Existing gait analyses for biometrics have mainly been conducted using video sequences [9]. However, such approaches require the user to be the only individual in front of the camera, and their accuracy may vary depending on the relative position of the camera.…”