Walker-Independent Features for Gait Recognition from Motion Capture Data

  title={Walker-Independent Features for Gait Recognition from Motion Capture Data},
  author={Michal Balazia and Petr Sojka},
MoCap-based human identification, as a pattern recognition discipline, can be optimized using a machine learning approach. Yet in some applications such as video surveillance new identities can appear on the fly and labeled data for all encountered people may not always be available. This work introduces the concept of learning walker-independent gait features directly from raw joint coordinates by a modification of the Fisher’s Linear Discriminant Analysis with Maximum Margin Criterion. Our… 

Gait Recognition from Motion Capture Data

Experiments on the CMU MoCap database show that the suggested method outperforms 13 relevant methods based on geometric features and a method to learn the features by a combination of Principal Component Analysis and Linear Discriminant Analysis.

You are how you walk: Uncooperative MoCap gait identification for video surveillance with incomplete and noisy data

It is shown that only a few examples of human gait cycles are required to learn a projection of raw MoCap data onto a low-dimensional subspace where the identities are well separable and Latent features learned by Maximum Margin Criterion (MMC) method discriminate better than any collection of geometric features.

An Evaluation Framework and Database for MoCap-Based Gait Recognition Methods

A framework and database to improve the development, evaluation and comparison of methods for gait recognition from Motion Capture (MoCap) data and contains a tool for learning a custom classifier and for classifying a custom query on a custom gallery.

Annotation Contribution to Classification Accuracy of Person Identification Based Gait Biometric

Accuration gap analysis shows that comparation between BD and RR shows no consistency rate on each amount of markers usage, and result shows that BD is always outperform to no annotation, while RR has lower accuracy to no annotations at using 10 markers.

Generating Virtual Avatars with Personalized Walking Gaits using Commodity Hardware

The algorithm is based on a perceptual study that evaluates self-recognition and similarity of gaits rendered on virtual avatars and proposes a data-driven synthesis algorithm that can generate a set of similar gaits from a single walker.

Target spike patterns enable efficient and biologically plausible learning for complex temporal tasks

A novel target-based learning scheme in which the learning rule derived from likelihood maximization is used to mimic a specific spatio-temporal spike pattern that encodes the solution to complex temporal tasks, making the learning extremely rapid and precise, outperforming state of the art algorithms for RSNNs.

Likelihood Based Learning Rule for Temporal Coding In Recurrent Spiking Neural Networks

This work proposes a novel target-based learning scheme in which an online approximation of the gradient ascent, in addition to guaranteeing complete locality in time and space, allows learning after very few presentations of the target output.

Digitization and Visualization of Folk Dances in Cultural Heritage: A Review

According to UNESCO, cultural heritage does not only include monuments and collections of objects, but also contains traditions or living expressions inherited from our ancestors and passed to our



Learning robust features for gait recognition by Maximum Margin Criterion

Experiments show that this method outperforms eight other relevant methods in terms of the distribution of biometric templates in respective feature spaces expressed in four class separability coefficients and is a leading concept for rank-based classifier systems.

Automatic Learning of Gait Signatures for People Identification

The experimental results indicate that using spatio-temporal cuboids of optical flow as input data for CNN allows to obtain state-of-the-art results on the gait task with an image resolution eight times lower than the previously reported results.

Gait Recognition with Kinect

An approach for gait recognition based on Microsoft Kinect, a peripheral for the gaming console XBOX 360, with an integrated depth sensor alowing for skeleton detection and tracking in realtime is presented.

View-invariant gait recognition via deterministic learning

  • Wei ZengCong Wang
  • Computer Science, Engineering
    2014 International Joint Conference on Neural Networks (IJCNN)
  • 2014
A new method to eliminate the effect of view angle for efficient gait recognition via deterministic learning theory is presented and comprehensive experiments are carried out on the CASIA-B and CMU gait databases to demonstrate the effectiveness of the proposed approach.

Applying Geometric Function on Sensors 3D Gait Data for Human Identification

This paper aims to propose a novel approach of human identification, which based on sensor data acquired by an optical system, has a unique relationship among the 3D human gaits, and uses this signature as to classify the human identification.

Person Identification Using Anthropometric and Gait Data from Kinect Sensor

This paper reports on experiments using a novel data set composed of 140 individuals walking in front of a Microsoft Kinect sensor, providing a methodology to extract anthropometric and gait features from this data and results of applying different machine learning algorithms on subject identification tasks.

Evaluation of different feature sets for gait recognition using skeletal data from Kinect

This paper construct and evaluate feature sets with the purpose of finding out the role of different types of features and body parts in the recognition process, and gives conclusions on which groups of featuresand body parts gave the best recognition rates.

Genetic feature selection for gait recognition

Motivated by the previous work on feature subset selection using genetic algorithms (GAs), this work proposes using GAs to select an optimal subset of gait features to improve gait recognition performance.

Person Identification using Skeleton Information from Kinect

Experimental results indicate that the proposed approach of simultaneous feature selection and classification is having better recognition accuracy compared to the earlier reported ones.

Person Identification in Natural Static Postures Using Kinect

This paper investigates the possibility of identification using static postures namely sitting and standing and indicates an identification accuracy of more than 97 % in terms of F-score for 10 people using a dataset created with various poses of natural Sitting and standing posture.