Ensemble of Heterogeneous Base Classifiers for Human Gait Recognition

Abstract

Human gait recognition is one of the most interesting issues within the subject of behavioral biometrics. The most significant problems connected with the practical application of biometric systems include their accuracy as well as the speed at which they operate, understood both as the time needed to recognize a particular person as well as the time necessary to create and train a biometric system. The present study made use of an ensemble of heterogeneous base classifiers to address these issues. A Heterogeneous ensemble is a group of classification models trained using various algorithms and combined to output an effective recognition A group of parameters identified on the basis of ground reaction forces was accepted as input signals. The proposed solution was tested on a sample of 322 people (5980 gait cycles). Results concerning the accuracy of recognition (meaning the Correct Classification Rate quality at 99.65%), as well as operation time (meaning the time of model construction at <12.5 min and the time needed to recognize a person at <0.1 s), should be considered as very good and exceed in quality other methods so far described in the literature.

Deep Metric Learning for Scalable Gait-Based Person Re-Identification Using Force Platform Data

“…When compared to other studies on force platform-based re-ID, the present benchmark was considerably higher than the previous benchmark at many-shot re-ID conducted across measurement sessions (86% on 79 IDs), yet lower than the previous benchmarks for many-shot re-ID conducted within a single session (99%+) [ 18 , 19 , 20 , 39 ]. When compared to vision-based re-ID systems, the present benchmark was notably higher than the accuracy of the current state-of-the-art system on CASIA-B under the most comparable conditions (82%) [ 27 ].…”

“…It should be noted that almost all studies in this field have utilized private datasets with fewer than 80 IDs [ 6 , 17 , 19 , 20 , 24 ]. Exceptions to this are Horst et al [ 18 ], who used data from 57 IDs that is available in in their public database [ 41 ], and Derlatka and Borowska [ 39 ], who used data from 322 IDs (that may be from their public dataset [ 42 ], though it was not mentioned). This makes ForceID A the second largest dataset to be implemented for force platform-based re-ID and by far the largest dataset of repeated measures over separate days.…”

“…The model obtained 86% validation accuracy when trained on samples from session one and then validated on samples from session two. Instead of using DNNs, Derlatka and Borowska [ 39 ] proposed an ensemble of simple machine learning classification models. This system obtained 100% accuracy on 322 IDs, who completed a single session of walking at preferred speed in personal footwear.…”

Human Gait Recognition for Biometrics Application Based on Deep Learning Fusion Assisted Framework

“…You may summarize the scenario using Eq. ( 10): y t+1 j = y prey − F y prey − y t j (10) Step three: a gentle assault with sharp dives (|E| ≥ 0.5, random () < 0.5). This search method employs a Levy flight-style search of the region of interest.…”

“…Gait recognition has several advantages over other biometric identifiers [9]. First, gait can be recorded from a distance without the participant's cooperation [10].…”