Gait Sequences Research Articles

Gait Analysis Using Computer Vision Based on Cloud Platform and Mobile Device

Frailty and senility are syndromes that affect elderly people. The ageing process involves a decay of cognitive and motor functions which often produce an impact on the quality of life of elderly people. Some studies have linked this deterioration of cognitive and motor function to gait patterns. Thus, gait analysis can be a powerful tool to assess frailty and senility syndromes. In this paper, we propose a vision-based gait analysis approach performed on a smartphone with cloud computing assistance. Gait sequences recorded by a smartphone camera are processed by the smartphone itself to obtain spatiotemporal features. These features are uploaded onto the cloud in order to analyse and compare them to a stored database to render a diagnostic. The feature extraction method presented can work with both frontal and sagittal gait sequences although the sagittal view provides a better classification since an accuracy of 95% can be obtained.

Multi-View Gait Recognition Based on a Spatial-Temporal Deep Neural Network

This paper proposes a novel spatial–temporal deep neural network (STDNN) that is applied to multi-view gait recognition. The STDNN comprises a temporal feature network (TFN) and a spatial feature network (SFN). In TFN, a feature sub-network is adopted to extract the low-level edge features of gait silhouettes. These features are input to the spatial–temporal gradient (STG) network that adopts a STG unit and a long short-term memory unit to extract the STG features. In SFN, the spatial features of gait sequences are extracted by multilayer convolutional neural networks from a gait energy image. The SFN is optimized by classification loss and verification loss jointly, which makes inter-class variations larger than intra-class variations. After training, the TFN and the SFN are employed to extract temporal and spatial features, respectively, which are applied to multi-view gait recognition. Finally, the combined predicted probability is adopted to identify individuals by the differences in their gaits. To evaluate the performance of the STDNN, extensive evaluations are carried out based on the CASIA-B, OU-ISIR, and CMU MoBo data sets. The best recognition scores achieved by STDNN are 95.67% under an identical view, 93.64% under a cross-view, and 92.54% under a multi-view. State-of-the-art approaches are compared with the STDNN in various situations. The results show that the STDNN outperforms the other methods and demonstrates the great potential of the STDNN for practical applications in the future.

Wearable sensors objectively measure gait parameters in Parkinson's disease.

Distinct gait characteristics like short steps and shuffling gait are prototypical signs commonly observed in Parkinson’s disease. Routinely assessed by observation through clinicians, gait is rated as part of categorical clinical scores. There is an increasing need to provide quantitative measurements of gait, e.g. to provide detailed information about disease progression. Recently, we developed a wearable sensor-based gait analysis system as diagnostic tool that objectively assesses gait parameter in Parkinson’s disease without the need of having a specialized gait laboratory. This system consists of inertial sensor units attached laterally to both shoes. The computed target of measures are spatiotemporal gait parameters including stride length and time, stance phase time, heel-strike and toe-off angle, toe clearance, and inter-stride variation from gait sequences. To translate this prototype into medical care, we conducted a cross-sectional study including 190 Parkinson’s disease patients and 101 age-matched controls and measured gait characteristics during a 4x10 meter walk at the subjects’ preferred speed. To determine intraindividual changes in gait, we monitored the gait characteristics of 63 patients longitudinally. Cross-sectional analysis revealed distinct spatiotemporal gait parameter differences reflecting typical Parkinson’s disease gait characteristics including short steps, shuffling gait, and postural instability specific for different disease stages and levels of motor impairment. The longitudinal analysis revealed that gait parameters were sensitive to changes by mirroring the progressive nature of Parkinson’s disease and corresponded to physician ratings. Taken together, we successfully show that wearable sensor-based gait analysis reaches clinical applicability providing a high biomechanical resolution for gait impairment in Parkinson’s disease. These data demonstrate the feasibility and applicability of objective wearable sensor-based gait measurement in Parkinson’s disease reaching high technological readiness levels for both, large scale clinical studies and individual patient care.

Gait-based human recognition using partial wavelet coherence and phase features

In this paper, a multi-view human gait recognition method which utilizes Partial Wavelet Coherence (PWC) as a novel feature is proposed. The Euclidean distance representation of PWC of the 1D signals generated due to movements of hands, legs, shoulders from multi-view gait sequences preserves the spatio-temporal information of walking individual. This method directly extracts the dynamic information without using any model. We got 73.26% average recognition accuracy when considered only PWC feature. Further, we investigate Phase Feature (PF) which also preserves discriminant information of dynamic phase angle between body parts. When PF considered additionally with PWC feature the system performance improved significantly and 82.52% average recognition accuracy reported.

Segmentation of gait sequences using inertial sensor data in hereditary spastic paraplegia.

Gait analysis is an important tool for diagnosis, monitoring and treatment of neurological diseases. Among these are hereditary spastic paraplegias (HSPs) whose main characteristic is heterogeneous gait disturbance. So far HSP gait has been analysed in a limited number of studies, and within a laboratory set up only. Although the rarity of orphan diseases often limits larger scale studies, the investigation of these diseases is still important, not only to the affect population, but also for other diseases which share gait characteristics.

Combining weighted adaptive CS-LBP and local linear discriminant projection for gait recognition

With the increasing demands of the remote surveillance system, the gait based personal identification research has obtained more and more attention from biometric recognition researchers. The gait sequence is easier to be affected by factors than other biometric feathers. In order to achieve better performance of the gait based identification system, in the paper, a local discriminant gait recognition method is proposed by integrating weighted adaptive center symmetric local binary pattern (WACS-LBP) with local linear discriminate projection (LLDP). The proposed method consists of two stages. In the first stage, the robust local weighted histogram feature vector is extracted from each gait image by WACS-LBP. In the second stage, the dimensionality of the extracted feature vector is reduced by LLDP. The highlights of the proposed method are (1) the extracted feature is robust to rotation invariant, and is also tolerant to illumination and pose changes; (2) the low dimensional feature vector reduced by LLDP can preserve the discriminating ability; and (3) the small-sample-size (SSS) problem is avoided naturally. The proposed method is validated and compared with the existing algorithms on a public gait database. The experimental results show that the proposed method is not only effective, but also can be clearly interpreted.

Human identification based on temporal lifting using 5/3 wavelet filters and radon transform

In this paper, a spatio-temporal gait recognition system is proposed to overcome the limitations associated with the most existing temporal template approaches such as gait energy image (GEI). These approaches do not preserve the whole temporal information in a gait sequence. They are also sensitive to changes in various conditions such as carrying and clothing. These limitations influence the performance of any gait recognition system. To address this problem, a temporal template approach based on lifting 5/3 wavelet filters is presented. In the proposed method named 5/3 gait image (5/3GI), the contour is first extracted from each image in a gait sequence. The gait contour images are then decomposed using 5/3 temporal wavelet filters into two temporal templates at the last temporal decomposition stage. These two templates are subjected to Radon transform for feature extraction. The principal component analysis (PCA) is subsequently applied to the Radon templates in the reference database to identify a subset of Radon template coefficients that carry the most important information suitable for gait recognition. Experimental results on USF HumanID and CASIA gait databases demonstrate that the proposed method achieves a better recognition performance than the most existing methods in the literature especially under walking variations.

Interlimb coordination during climbing by primates: a test of the Vilensky hypothesis

Primate horizontal locomotion is characterized by a trio of unusual features, including hindlimb‐biased weight support, greater diagonality, and high degrees of forelimb protraction. All of these features are thought to have been influenced by early primates' adoption of a thin‐branch niche. This interpretation has however been challenged based on non‐primate mammals that occupy a similar niche, but do not demonstrate these characteristics. Vilensky (1994) proposed that diagonality evolved in response to greater use of more vertically oriented supports. Nyakatura et al (2008) provided data on tamarins in support of this hypothesis. However, both of these studies examined limited primate species and only on inclined supports, rather than purely vertical supports. Here, we examine diagonality during climbing of vertical supports compared to walking across horizontal supports to the hypothesis first proposed by Vilensky.711 symmetric gait sequences were collected during horizontal walking and vertical climbing by 8 species of primates spanning over an order of magnitude of body weight. Trials were video recorded and footfall sequence were determined for each trial from these recording. From these data, speed, diagonality, interlimb coordination, forelimb and hindlimb duty factor, and handhold patterns were determined and compared between horizontal walking and vertical climbing.During horizontal quadrupedal locomotion, most primates tended to utilize diagonal‐sequence diagonal‐couplet gaits and trots. These gait types involve the contralateral forelimbs and hindlimbs moving together in near‐unison, and therefore limit the portion of the stride that the limbs are arranged as a unilateral bipod. In contrast, during climbing all species demonstrated decreased diagonality (ANCOVA, p<0.0001), thereby increasing the proportion of trots and lateral‐sequence lateral couplet gaits. The degree of this decreased diagonality appears to be species specific, with shorter limbed species demonstrating a greater decrease than longer limbed species. The amount the limbs are in contact with the support also increases during climbing, thereby increasing the number of limbs that are in contact with the support at any one time.Although most primates exhibit primarily diagonal sequence, diagonal couplet gaits during horizontal walking, it appears that climbing does not follow a similar pattern. Diagonality does not increase on vertical supports, as suggested by Vilensky and supported by Nyakatura, although diagonal couplets are retained. Our results are similar to those recently reported by Schoonaert et al (2016) for bonobos. Additionally, duty factor index appears to be increased during climbing. We suggest that these phenomena are due to the propulsive requirements of vertical climbing, which causes hindlimb duty factor to increase disproportionately relative to the forelimb in some primates.Support or Funding InformationThis project was supported by the National Science Foundation, Leakey Foundation, Force and Motion Foundation, National Science Foundation's Graduate Research Fellowship Program, Molly Glander award, Department of Veterinary Sciences at the Michale E. Keeling Center at UT MD Anderson Cancer Center, and Duke University Internal Funding.

Optimal three-dimensional biped walking pattern generation based on geodesics

The innovative three-dimensional humanoid biped gait planning method using geodesics is introduced in this article. In order to control three-dimensional walking, the three-dimensional linear inverted pendulum model is studied in our energy-optimal gait planning based on geodesics. The kinetic energy of the three-dimensional linear inverted pendulum model is calculated at first. Based on this kinetic energy model, the Riemannian metric is defined and the Riemannian surface is further determined by this Riemannian metric. The geodesic is the shortest line between two points on the Riemannian surface. This geodesic is the optimal kinetic energy gait for the center of gravity because the kinetic energy along the geodesic is invariant according to the geometric property of geodesics and the walking is energy-saving. Finally, a simulation experiment using a 12-degree-of-freedom biped robot model is implemented. The gait sequences of the simulated RoboErectus humanoid robot are obtained in the ROS (Robot Operating System) Gazebo environment. The proposed geodesics approach is compared with the traditional sinusoidal interpolation method by analyzing the torque feedback of each joint of both legs, and our geodesics optimization gait planning method for three-dimensional linear inverted pendulum model walking control is verified by the assessment results.

Simultaneous Contact, Gait and Motion Planning for Robust Multi-Legged Locomotion via Mixed-Integer Convex Optimization

Traditional motion planning approaches for multi-legged locomotion divide the problem into several stages, such as contact search and trajectory generation. However, reasoning about contacts and motions simultaneously is crucial for the generation of complex whole-body behaviors. Currently, coupling theses problems has required either the assumption of a fixed gait sequence and flat terrain condition, or non-convex optimization with intractable computation time. In this paper, we propose a mixed-integer convex formulation to plan simultaneously contact locations, gait transitions and motion, in a computationally efficient fashion. In contrast to previous works, our approach is not limited to flat terrain nor to a pre-specified gait sequence. Instead, we incorporate the friction cone stability margin, approximate the robot's torque limits, and plan the gait using mixed-integer convex constraints. We experimentally validated our approach on the HyQ robot by traversing different challenging terrains, where non-convexity and flat terrain assumptions might lead to sub-optimal or unstable plans. Our method increases the motion generality while keeping a low computation time.

Pose-based gait recognition with local gradient descriptors and hierarchically aggregated residuals

We focus on the problem of pose-based gait recognition. Our contribution is two-fold. First, we incorporate a local histogram descriptor that allows us to encode the trajectories of selected limbs via a one-dimensional version of histogram of oriented gradients features. In this way, a gait sequence is encoded into a sequence of local gradient descriptors. Second, we utilize a robust encoding method in which the residuals of local descriptors, with respect to a discriminative model, are aggregated into fixed length vectors. This technique combines the advantages of both residual aggregation and soft-assignment techniques, resulting in a powerful vector representation. For classification purposes, we use a nonlinear kernel to map vectors into a reproducing kernel Hilbert space. Then, we classify an encoded gait sequence according to the sparse representation-based classification method. Experimental evaluation on two publicly available datasets demonstrates the effectiveness of the proposed scheme on both recognition and verification tasks.

Design of adaptive biometric gait recognition algorithm with free walking directions

Gait is one of well-identified biometrics that has been broadly applied for human identification at a distance based on their motion style. However, the current gait recognition might have difficulties due to changing the viewing angles anduncertainty associated with gait signature extraction. This study deals with the design of an intelligent gait recognition system that tackles the problems mentioned above. This system is based on spatial-domain energy deviation image as a gait signature by adopting clustering technique to estimate the gait period in the gait sequence with arbitrary walking directions. To further improve the performance of the proposed system, interval type-2 fuzzy K-nearest neighbor classifier is used to diminish the effect of uncertainty formed by variations in gait signature extraction. Interval type-2 fuzzy set is involved in extending themembership values of each gait signature by using several initial K in order to handle and manage uncertainty that exists in choosing the initial value K. The proposed method realises the reduction in the dimensions of the gait feature and over-fitting. The comprehensive analyses reveal that the proposed algorithm can significantly enhance the multiple view gait recognition performance when being matched to the similar methods in the literature.

Gait based recognition via fusing information from Euclidean and Riemannian manifolds

Gait is a particular periodical type of human motion with several unique characteristics for every person. In this work we focus on the problem of pose based gait recognition. The contribution of the proposed work is threefold. First we represent every gait sequence according to both the deviation of the poses from an appropriate global model, as well as the intra-sequence pose variability. Secondly, we propose a method which allows us to fuse information from feature representations from both Euclidean and Riemannian spaces by mapping data in a Reproducing Kernel Hilbert Space (RKHS). Classification is then performed using a kernelized version of the SRC algorithm. Third we present a new publicly available dataset for pose based gait recognition captured with Kinect V2. Experimental evaluation reveals state-of-the-art performance in both recognition and verification tasks and a capacity for real-time operation.

POSTURAL CLIMBING BEHAVIOUR OF DIDELPHID MARSUPIALS: PARALLELS WITH PRIMATES

Differences in body size and in the use of arboreal strata limit the climbing behaviour and performance of didelphids. Similarly to primates, the arboreal canopy dweller, Caluromys philander exhibits a diagonal-sequence gait. In contrast, terrestrial didelphids use a symmetrical lateral sequence in horizontal locomotion. Postural behaviour along thin supports may allow understanding the mechanisms behind the higher performance of arboreal didelphids climbing. Here in, we describe and compare gait sequence and postural behaviour of seven didelphids climbing a slender and flexible vertical support. Animals were stimulated to climb a vertical support of 1.25 cm diameter. Postural behaviour was qualitatively described for each species in the cycle of maximum velocity by a frame-by-frame analysis of gait cycles, evaluating (1) movements of the tail, (2) posture of hand and wrist when grasping, (3) distance of the body to the support, (4) lateral swinging of the body, (5) orientation of the head, (6) limb posture, and (7) stride cycle. Only arboreal species were capable of climbing with only two limbs grasping the support, keeping a straight body orientation at some distance from the support, and sustaining a more constant and regular climbing velocity. The tail played a role in didelphids with better climbing performance (i.e., higher relative velocity), counteracting the lateral swinging of the body, helping with the animal balance. However, the prehensile ability of the tail was not used in climbing. The most stunning result is that all didelphids grasped the rope between digits 2-3, a schizaxonic grasp, involving a neutral hand orientation regarding the ulna. The didelphids protracted the humerus at forelimb touchdown, and the angle between the arm and the horizontal body axis was greater than 90°. The same postures were already observed in horizontal locomotion of C. philander, Monodelphis domestica, and primates. Locomotory and postural adaptations for an arboreal lifestyle in didelphids seem to be limited to small to medium body sizes, up to the size of species of Caluromys. The arboreal locomotion of didelphids is an important key to understand adaptation and evolution of mammals to an arboreal niche, and the comparison with small primates may help to identify adaptive convergence to arboreal locomotion.

Tensorial Feed Forward Neural Networks with Random Weights for Gait Recognition Using MPCA Features

Because of gait sequences are naturally three-dimensional data, there have been several tensorial feature extraction methods to deal with tensors while there are effective tensorial classifiers. In this work, by using a linear tensor projection, a new classifier based on neural networks with random weights is introduced. Due to the proposed algorithm can classify gait samples directly without vectorizing them, the intrinsic structure information of the input data can be reserved. In addition, discriminative features sets are generated using MPCA to ascertain classification accuracy. Finally, Extensive experiments are carried out on two gait databases and results are compared against state-of-the-art techniques. It is demonstrated that the proposed algorithm MPCA plus TNNRE achieves better recognition performance.

Automated classification of pathological gait after stroke using ubiquitous sensing technology.

This study uses machine learning methods to distinguish between healthy and pathological gait. Examples of multi-dimensional pathological and normal gait sequences were collected from post-stroke and healthy individuals in a real clinical setting and with two Kinect sensors. The trajectories of rotational angle and global velocity of selected body joints (hips, spine, shoulders, neck, knees and ankles) over time formed the gait sequences. The combination of k nearest neighbor (kNN) and dynamic time warping (DTW) was used for classification. Leave one subject out cross validation was implemented to evaluate the performance of the binary classifier in terms of F1-score in the original feature space, and also in a reduced dimensional feature space using PCA. The pair of k = 1 in kNN and the warping window size 25% of gait sequences in DTW achieved maximum F1-score. Using PCA, pathological gait sequences were discriminated from healthy sequences with the F1-score = 96%.

Gait recognition from corrupted silhouettes: a robust statistical approach

This paper introduces a method based on robust statistics to build reliable gait signatures from averaging silhouette descriptions, mainly when gait sequences are affected by severe and persistent defects. The term robust refers to the ability of reducing the impact of silhouette defects (outliers) on the average gait pattern, while taking advantage of clean silhouette regions. An extensive experimental framework was defined based on injecting three types of realistic defects (salt and pepper noise, static occlusion, and dynamic occlusion) to clean gait sequences, both separately in an easy setting and jointly in a hard setting. The robust approach was compared against two other operation modes: (1) simple mean (weak baseline) and (2) defect exclusion (strong benchmark). Three gait representation methods based on silhouette averaging were used: Gait Energy Image (GEI), Gradient Histogram Energy Image (GHEI), and the joint use of GEI and HOG descriptors. Quality of gait signatures was assessed by their discriminant power in a large number of gait recognition tasks. Nonparametric statistical tests were applied on recognition results, searching for significant differences between operation modes.

Fusion of sparse representation and dictionary matching for identification of humans in uncontrolled environment

Biomechanics based human identification is a major area of research. Biomechanics based approaches depend on accurately recognizing humans using body movements, the accuracy of these approaches is enhanced by incorporating the knee-hip angle to angle relationships. Current biomechanics based models are developed by considering the biomechanics of human walking and running. In biomechanics the joint angle characteristics, also known as gait features play a vital role in identification of humans. In general, identification of humans can be broadly classified into two approaches: biomechanics based approach, also known as Gait Recognition and biometric based Composite Sketch Matching. Gait recognition is a biomechanics based approach which uses gait traits for person authentication, it discriminates people by the way they walk.Gait recognition uses shape and motion information of a person and identifies the individual; this information is generally acquired from an image sequence. The efficiency of gait recognition is mainly affected by covariates such as observation view, walking speed, clothing, and belongings. Biometric based approach for human identification is usually done by composite sketch matching. Composite sketches are sketches generated using a computer. This obviates the need of using a skilled sketch artist; these sketches can be easily drawn by eyewitness using face design system software in a very short time period. This doesn’t require any prior specialized software training but identifying humans using only composite sketches is still a challenging task owing to the fact that human faces are not always clearly visible from a distance. Hence drawing a composite sketch at all times is not feasible.The key contribution of this paper is a fusion system developed by combining biomechanics based gait recognition and biometric based composite sketch matching for identifying humans in crowded scenes. First various existing biomechanics based approaches for gait recognitionare developed. Then a novel biomechanics based gait recognition is developed using Sparse Representation to generate what we term as “score 1.” Further another novel technique for composite sketch matching is developed using Dictionary Matching to generate what we term as “score 2.” Finally, score level fusion using Dempster Shafer and Proportional Conflict Distribution Rule Number 5 is performed. The proposed fusion approach is validated using a database containing biomechanics based gait sequences and biometric based composite sketches. From our analysis we find that a fusion of gait recognition and composite sketch matching provides excellent results for real-time human identification.

Gait characteristics and spatio-temporal variables of climbing in bonobos (Pan paniscus).

Although much is known about the terrestrial locomotion of great apes, their arboreal locomotion has been studied less extensively. This study investigates arboreal locomotion in bonobos (Pan paniscus), focusing on the gait characteristics and spatio-temporal variables associated with locomotion on a pole. These features are compared across different substrate inclinations (0°, 30°, 45°, 60°, and 90°), and horizontal quadrupedal walking is compared between an arboreal and a terrestrial substrate. Our results show greater variation in footfall patterns with increasing incline, resulting in more lateral gait sequences. During climbing on arboreal inclines, smaller steps and strides but higher stride frequencies and duty factors are found compared to horizontal arboreal walking. This may facilitate better balance control and dynamic stability on the arboreal substrate. We found no gradual change in spatio-temporal variables with increasing incline; instead, the results for all inclines were clustered together. Bonobos take larger strides at lower stride frequencies and lower duty factors on a horizontal arboreal substrate than on a flat terrestrial substrate. We suggest that these changes are the result of the better grip of the grasping feet on an arboreal substrate. Speed modulation of the spatio-temporal variables is similar across substrate inclinations and between substrate types, suggesting a comparable underlying motor control. Finally, we contrast these variables of arboreal inclined climbing with those of terrestrial bipedal locomotion, and briefly discuss the results with respect to the origin of habitual bipedalism. Am. J. Primatol. 78:1165-1177, 2016. © 2016 Wiley Periodicals, Inc.

Clothing and carrying condition invariant gait recognition based on rotation forest

This paper proposes a gait recognition method which is invariant to maximum number of challenging factors of gait recognition mainly unpredictable variation in clothing and carrying conditions. The method introduces an averaged gait key-phase image (AGKI) which is computed by averaging each of the five key-phases of the gait periods of a gait sequence. It analyses the AGKIs using high-pass and low-pass Gaussian filters, each at three cut-off frequencies to achieve robustness against unpredictable variation in clothing and carrying conditions in addition to other covariate factors, e.g., walking speed, segmentation noise, shadows under feet and change in hair style and ground surface. The optimal cut-off frequencies of the Gaussian filters are determined based on an analysis of the focus values of filtered human subject’s silhouettes. The method applies rotation forest ensemble learning recognition to enhance both individual accuracy and diversity within the ensemble for improved identification rate. Extensive experiments on public datasets demonstrate the efficacy of the proposed method.