Robust Inertial Motion Tracking through Deep Sensor Fusion across Smart Earbuds and Smartphone

2021 ◽  
Vol 5 (2) ◽  
pp. 1-26
Author(s):  
Jian Gong ◽  
Xinyu Zhang ◽  
Yuanjun Huang ◽  
Ju Ren ◽  
Yaoxue Zhang
Keyword(s):  
Sensor Fusion ◽  
Motion Tracking ◽  
Large Scale ◽  
Body Movement ◽  
Motion Artifacts ◽  
Indoor Navigation ◽  
The Body ◽  
Data Driven ◽  
Whole Body ◽  
Attention Model

IMU based inertial tracking plays an indispensable role in many mobility centric tasks, such as robotic control, indoor navigation and virtual reality gaming. Despite its mature application in rigid machine mobility (e.g., robot and aircraft), tracking human users via mobile devices remains a fundamental challenge due to the intractable gait/posture patterns. Recent data-driven models have tackled sensor drifting, one key issue that plagues inertial tracking. However, these systems still assume the devices are held or attached to the user body with a relatively fixed posture. In practice, natural body activities may rotate/translate the device which may be mistaken as whole body movement. Such motion artifacts remain as the dominating factor that fails existing inertial tracing systems in practical uncontrolled settings. Inspired by the observation that human heads induces far less intensive movement relative to the body during walking, compared to other parts, we propose a novel multi-stage sensor fusion pipeline called DeepIT, which realizes inertial tracking by synthesizing the IMU measurements from a smartphone and an associated earbud. DeepIT introduces a data-driven reliability aware attention model, which assesses the reliability of each IMU and opportunistically synthesizes their data to mitigate the impacts of motion noise. Furthermore, DeepIT uses a reliability aware magnetometer compensation scheme to combat the angular drifting problem caused by unrestricted motion artifacts. We validate DeepIT on the first large-scale inertial navigation dataset involving both smartphone and earbud IMUs. The evaluation results show that DeepIT achieves multiple folds of accuracy improvement on the challenging uncontrolled natural walking scenarios, compared with state-of-the-art closed-form and data-driven models.

Tuning of a Basic Coordination Pattern Constructs Straight-Ahead and Curved Walking in Humans

2004 ◽  
Vol 91 (4) ◽  
pp. 1524-1535 ◽  
Author(s):  
Grégoire Courtine ◽  
Marco Schieppati
Keyword(s):  
Principal Components ◽  
Lower Limb ◽  
Time Course ◽  
Body Movement ◽  
The Body ◽  
Whole Body ◽  
Coordination Pattern ◽  
Data Set ◽  
Coordination Patterns ◽  
Straight Ahead

We tested the hypothesis that common principles govern the production of the locomotor patterns for both straight-ahead and curved walking. Whole body movement recordings showed that continuous curved walking implies substantial, limb-specific changes in numerous gait descriptors. Principal component analysis (PCA) was used to uncover the spatiotemporal structure of coordination among lower limb segments. PCA revealed that the same kinematic law accounted for the coordination among lower limb segments during both straight-ahead and curved walking, in both the frontal and sagittal planes: turn-related changes in the complex behavior of the inner and outer limbs were captured in limb-specific adaptive tuning of coordination patterns. PCA was also performed on a data set including all elevation angles of limb segments and trunk, thus encompassing 13 degrees of freedom. The results showed that both straight-ahead and curved walking were low dimensional, given that 3 principal components accounted for more than 90% of data variance. Furthermore, the time course of the principal components was unchanged by curved walking, thereby indicating invariant coordination patterns among all body segments during straight-ahead and curved walking. Nevertheless, limb- and turn-dependent tuning of the coordination patterns encoded the adaptations of the limb kinematics to the actual direction of the walking body. Absence of vision had no significant effect on the intersegmental coordination during either straight-ahead or curved walking. Our findings indicate that kinematic laws, probably emerging from the interaction of spinal neural networks and mechanical oscillators, subserve the production of both straight-ahead and curved walking. During locomotion, the descending command tunes basic spinal networks so as to produce the changes in amplitude and phase relationships of the spinal output, sufficient to achieve the body turn.

Caudal Fin and Body Movement in the Propulsion of some Fish

1963 ◽  
Vol 40 (1) ◽  
pp. 23-56 ◽  
Author(s):  
RICHARD BAINBRIDGE
Keyword(s):  
Body Movement ◽  
Wave Length ◽  
The Body ◽  
Whole Body ◽  
Tail Region ◽  
Caudal Fin ◽  
A Value ◽  
Total Thrust ◽  
Transverse Movement ◽  
Relationship Of

1. Observations made on bream, goldfish and dace swimming in the ‘Fish Wheel’ apparatus are described. These include: 2. An account of the complex changes in curvature of the caudal fin during different phases of the normal locomotory cycle. Measurements of this curvature and of the angles of attack associated with it are given. 3. An account of changes in area of the caudal fin during the cycle of lateral oscillation. Detailed measurements of these changes, which may involve a 30 % increase in height or a 20 % increase in area, are given. 4. An account of the varying speed of transverse movement of the caudal fin under various conditions and the relationship of this to the changes in area and amount of bending. Details of the way this transverse speed may be asymmetrically distributed relative to the axis of progression of the fish are given. 5. An account of the extent of the lateral propulsive movements in other parts of the body. These are markedly different in the different species studied. Measurements of the wave length of this movement and of the rate of progression of the wave down the body are given. 6. It is concluded that the fish has active control over the speed, the amount of bending and the area of the caudal fin during transverse movement. 7. The bending of the fin and its changes in area are considered to be directed to the end of smoothing out and making more uniform what would otherwise be an intermittent thrust from the oscillating tail region. 8. Some assessment is made of the proportion of the total thrust contributed by the caudal fin. This is found to vary considerably, according to the form of the lateral propulsive movements of the whole body, from a value of 45% for the bream to 84% for the dace.

Touching Space: Using Motion Capture and Stereo Projection to Create a “Virtual Haptics” of Dance

Leonardo ◽  
2010 ◽  
Vol 43 (4) ◽  
pp. 359-366 ◽  
Author(s):  
Kim Vincs ◽  
John McCormick
Keyword(s):  
Motion Capture ◽  
Large Scale ◽  
Body Movement ◽  
Whole Body ◽  
Time Motion ◽  
Computer Interfaces ◽  
Human Computer Interfaces ◽  
Virtual Force ◽  
Spatial Volume ◽  
Whole Body Movement

This paper describes the work of a group of artists in Australia who used real-time motion capture and 3D stereo projection to create a large-scale performance environment in which dancers seemed to “touch” the volume. This project re-versions Suzanne Langer's 1950s philosophy of dance as “virtual force” to realize the idea of a “virtual haptics” of dance that extends the dancer's physical agency literally across and through the surrounding spatial volume. The project presents a vision of interactive dance performance that “touches” space by visualizing kinematics as intentionality and agency. In doing so, we suggest the possibility of new kinds of human-computer interfaces that emphasize touch as embodied, nuanced agency that is mediated by the subtle qualities of whole-body movement, in addition to more goal-oriented, task-based gestures such as pointing or clicking.

scholarly journals Effect of Leg Dominance on The Center-of-Mass Kinematics During an Inside-of-the-Foot Kick in Amateur Soccer Players

2014 ◽  
Vol 42 (1) ◽  
pp. 51-61 ◽  
Author(s):  
Matteo Zago ◽  
Andrea Francesco Motta ◽  
Andrea Mapelli ◽  
Isabella Annoni ◽  
Christel Galvani ◽  
...  
Keyword(s):  
Body Movement ◽  
Center Of Mass ◽  
Three Dimensional ◽  
The Body ◽  
Soccer Players ◽  
Upper Body ◽  
Whole Body ◽  
Movement Velocity ◽  
Ground Support ◽  
Analysis System

Abstract Soccer kicking kinematics has received wide interest in literature. However, while the instep-kick has been broadly studied, only few researchers investigated the inside-of-the-foot kick, which is one of the most frequently performed techniques during games. In particular, little knowledge is available about differences in kinematics when kicking with the preferred and non-preferred leg. A motion analysis system recorded the three-dimensional coordinates of reflective markers placed upon the body of nine amateur soccer players (23.0 ± 2.1 years, BMI 22.2 ± 2.6 kg/m2), who performed 30 pass-kicks each, 15 with the preferred and 15 with the non-preferred leg. We investigated skill kinematics while maintaining a perspective on the complete picture of movement, looking for laterality related differences. The main focus was laid on: anatomical angles, contribution of upper limbs in kick biomechanics, kinematics of the body Center of Mass (CoM), which describes the whole body movement and is related to balance and stability. When kicking with the preferred leg, CoM displacement during the ground-support phase was 13% higher (p<0.001), normalized CoM height was 1.3% lower (p<0.001) and CoM velocity 10% higher (p<0.01); foot and shank velocities were about 5% higher (p<0.01); arms were more abducted (p<0.01); shoulders were rotated more towards the target (p<0.01, 6° mean orientation difference). We concluded that differences in motor control between preferred and non-preferred leg kicks exist, particularly in the movement velocity and upper body kinematics. Coaches can use these results to provide effective instructions to players in the learning process, moving their focus on kicking speed and upper body behavior

scholarly journals Food responsiveness regulates episodic behavioral states in Caenorhabditis elegans

2017 ◽  
Vol 117 (5) ◽  
pp. 1911-1934 ◽  
Author(s):  
Richard J. McCloskey ◽  
Anthony D. Fouad ◽  
Matthew A. Churgin ◽  
Christopher Fang-Yen
Keyword(s):  
Nervous System ◽  
Caenorhabditis Elegans ◽  
Body Movement ◽  
Egg Laying ◽  
The Body ◽  
Hierarchical Organization ◽  
Whole Body ◽  
Locomotory Activity ◽  
Behavioral States ◽  
External Environments

Animals optimize survival and reproduction in part through control of behavioral states, which depend on an organism’s internal and external environments. In the nematode Caenorhabditis elegans a variety of behavioral states have been described, including roaming, dwelling, quiescence, and episodic swimming. These states have been considered in isolation under varied experimental conditions, making it difficult to establish a unified picture of how they are regulated. Using long-term imaging, we examined C. elegans episodic behavioral states under varied mechanical and nutritional environments. We found that animals alternate between high-activity (active) and low-activity (sedentary) episodes in any mechanical environment, while the incidence of episodes and their behavioral composition depend on food levels. During active episodes, worms primarily roam, as characterized by continuous whole body movement. During sedentary episodes, animals exhibit dwelling (slower movements confined to the anterior half of the body) and quiescence (a complete lack of movement). Roaming, dwelling, and quiescent states are manifest not only through locomotory characteristics but also in pharyngeal pumping (feeding) and in egg-laying behaviors. Next, we analyzed the genetic basis of behavioral states. We found that modulation of behavioral states depends on neuropeptides and insulin-like signaling in the nervous system. Sensory neurons and the Foraging homolog EGL-4 regulate behavior through control of active/sedentary episodes. Optogenetic stimulation of dopaminergic and serotonergic neurons induced dwelling, implicating dopamine as a dwell-promoting neurotransmitter. Our findings provide a more unified description of behavioral states and suggest that perception of nutrition is a conserved mechanism for regulating animal behavior. NEW & NOTEWORTHY One strategy by which animals adapt to their internal states and external environments is by adopting behavioral states. The roundworm Caenorhabditis elegans is an attractive model for investigating how behavioral states are genetically and neuronally controlled. Here we describe the hierarchical organization of behavioral states characterized by locomotory activity, feeding, and egg-laying. We show that decisions to engage in these behaviors are controlled by the nervous system through insulin-like signaling and the perception of food.

scholarly journals Wearable Magnetic Induction-based Approach Toward 3D Motion Tracking

2021 ◽  
Author(s):  
Negar Golestani ◽  
Mahta Moghaddam
Keyword(s):  
Magnetic Induction ◽  
Motion Tracking ◽  
Mean Squared Error ◽  
Average Distance ◽  
Tracking System ◽  
Body Movement ◽  
The Body ◽  
Machine Learning Techniques ◽  
Wide Range ◽  
Battery Capacity

Abstract Activity recognition using wearable sensors has gained popularity due to its wide range of applications, including healthcare, rehabilitation, sports, and senior monitoring. Tracking the body movement in 3D space facilitates behavior recognition in different scenarios. Wearable systems have limited battery capacity, and many critical challenges have to be addressed to gain a trade-off among power consumption, computational complexity, minimizing the effects of environmental interference, and achieving higher tracking accuracy. This work presents a motion tracking system based on magnetic induction (MI) to tackle the challenges and limitations inherent in designing a wireless monitoring system. We integrated a realistic prototype of an MI sensor with machine learning techniques and investigated one-sensor and two-sensor configuration setups for motion reconstruction. This approach is successfully evaluated using measured and synthesized datasets generated by the analytical model of the MI system. The system has an average distance root-mean-squared error (RMSE) error of 3 cm compared to the ground-truth real-world measured data with Kinect.

scholarly journals Sociality and Interaction Envelope Organize Visual Action Representations

2019 ◽  
Author(s):  
Leyla Tarhan ◽  
Talia Konkle
Keyword(s):  
Visual Cortex ◽  
Large Scale ◽  
Functional Neuroimaging ◽  
Action Observation ◽  
The Body ◽  
Whole Body ◽  
Action Perception ◽  
Social Content ◽  
Wide Range ◽  
The Brain

Humans observe a wide range of actions in their surroundings. How is the visual cortex organized to process this diverse input? Using functional neuroimaging, we measured brain responses while participants viewed short videos of everyday actions, then probed the structure in these responses using voxel-wise encoding modeling. Responses were well fit by feature spaces that capture the body parts involved in an action and the action’s targets (i.e. whether the action was directed at an object, another person, the actor, and space). Clustering analyses revealed five large-scale networks that summarized the voxel tuning: one related to social aspects of an action, and four related to the scale of the interaction envelope, ranging from fine-scale manipulations directed at objects, to large-scale whole-body movements directed at distant locations. We propose that these networks reveal the major representational joints in how actions are processed by visual regions of the brain.Significance StatementHow does the brain perceive other people’s actions? Prior work has established that much of the visual cortex is active when observing others’ actions. However, this activity reflects a wide range of processes, from identifying a movement’s direction to recognizing its social content. We investigated how these diverse processes are organized within the visual cortex. We found that five networks respond during action observation: one that is involved in processing actions’ social content, and four that are involved in processing agent-object interactions and the scale of the effect that these actions have on the world (its “interaction envelope”). Based on these findings, we propose that sociality and interaction envelope size are two of the major features that organize action perception in the visual cortex.

scholarly journals Computation-Based Feature Representation of Body Expressions in the Human Brain

2020 ◽  
Vol 30 (12) ◽  
pp. 6376-6390
Author(s):  
Marta Poyo Solanas ◽  
Maarten Vaessen ◽  
Beatrice de Gelder
Keyword(s):  
Brain Activity ◽  
Action Observation ◽  
Body Movement ◽  
The Body ◽  
Feature Representation ◽  
Primate Species ◽  
Whole Body ◽  
Affective Neuroscience ◽  
Body Area ◽  
Extrastriate Body Area

Abstract Humans and other primate species are experts at recognizing body expressions. To understand the underlying perceptual mechanisms, we computed postural and kinematic features from affective whole-body movement videos and related them to brain processes. Using representational similarity and multivoxel pattern analyses, we showed systematic relations between computation-based body features and brain activity. Our results revealed that postural rather than kinematic features reflect the affective category of the body movements. The feature limb contraction showed a central contribution in fearful body expression perception, differentially represented in action observation, motor preparation, and affect coding regions, including the amygdala. The posterior superior temporal sulcus differentiated fearful from other affective categories using limb contraction rather than kinematics. The extrastriate body area and fusiform body area also showed greater tuning to postural features. The discovery of midlevel body feature encoding in the brain moves affective neuroscience beyond research on high-level emotion representations and provides insights in the perceptual features that possibly drive automatic emotion perception.

Changthangi Goats: A rich source of pashmina production in Ladakh

2004 ◽  
Vol 35 ◽  
pp. 75-85 ◽  
Author(s):  
T.K. Bhattacharya ◽  
S.S. Misra ◽  
F.D. Sheikh ◽  
P. Kumar ◽  
A. Sharma
Keyword(s):  
Growth Hormone ◽  
Large Scale ◽  
Fibre Diameter ◽  
The Body ◽  
Whole Body ◽  
Growth Hormone Gene ◽  
Body Colour ◽  
Male And Female ◽  
Pcr Rflp ◽  
Preliminary Study

SummaryPashmina, internationally known as “cashmere”, a fine luxury fibre, is being produced from Changthangi goats bred in the Ladakh region of India. The Leh district of Greater Ladakh produces around 30 000 kg of pashmina fibre which is harvested from about 0.15 million Changthangi goats reared by the Changpa nomads in Changthang region of Greater Ladakh. Changthangi goats are sometimes also called Changra goats. Pashmina producing goats are of great importance for revitalising the economy of the poverty stricken region of Changthang and the Leh district of Ladakh. The information on Changthangi goats was collected from both small and large-scale farmers in Ladakh, and 337 animals were included in the study.The body colour of Changthangi goats varies from white to light brown and nearly whole body is covered with pashmina and long hairs. The average birth weights of male and female kids were estimated as 2.11 ± 0.3 and 2.06 ± 0.2 kg, respectively while weight at 300 days was found to be 20.0 ± 2.1 and 18.7 ± 1.9 kg in male and females respectively.The pashmina yields of bucks, does, male hoggets and female hoggets were estimated as 402 ± 19 g, 248 ± 14 g, 255 ± 12 and 280 ± 16 g, respectively. The length of pashmina fibre was found to be 4.25 ± 1.2 cm in males and 4.02 ± 1.5 cm in females while fibre diameter in male and female goats was estimated as 12.9 ± 2.6 μ and 13.0 ± 3.0 μ, respectively.The twining rate was found to be very low, nearly 0.3%. A preliminary study at DNA level with PCR-RFLP indicated monomorphism at the growth hormone gene. The occurrence of disease was quite low although some genetic deformities in this breed were not uncommon.

scholarly journals The onset time of balance control during walking is phase-independent, but the magnitude of the response is not

2019 ◽  
Author(s):  
Hendrik Reimann ◽  
Tyler Fettrow ◽  
David Grenet ◽  
Elizabeth D. Thompson ◽  
John J. Jeka
Keyword(s):  
Nervous System ◽  
Gait Cycle ◽  
Center Of Pressure ◽  
Body Movement ◽  
Reaction Force ◽  
The Body ◽  
Upper Body ◽  
Whole Body ◽  
Lower Body ◽  
The Central Nervous System

AbstractThe human body is mechanically unstable during walking. Maintaining upright stability requires constant regulation of muscle force by the central nervous system to push against the ground and move the body mass in the desired way. Activation of muscles in the lower body in response to sensory or mechanical perturbations during walking is usually highly phase-dependent, because the effect any specific muscle force has on the body movement depends upon the body configuration. Yet the resulting movement patterns of the upper body after the same perturbations are largely phase-independent. This is puzzling, because any change of upper-body movement must be generated by parts of the lower body pushing against the ground. How do phase-dependent muscle activation patterns along the lower body generate phase-independent movement patterns of the upper body? We hypothesize that in response to a perceived threat to balance, the nervous system generates a functional response by pushing against the ground in any way possible with the current body configuration. This predicts that the changes in the ground reaction force patterns following a balance perturbation should be phase-independent. Here we test this hypothesis by disturbing upright balance using Galvanic vestibular stimulation at three different points in the gait cycle. We measure the resulting changes in whole-body center of mass movement and the location of the center of pressure of the ground reaction force. We find that the whole-body balance response is not phase-independent as expected: balance responses are initiated faster and are smaller following a disturbance late in the gait cycle. Somewhat paradoxically, the initial center of pressure changes are larger for perturbations late in the gait cycle. The onset of the center of pressure changes however, does not depend on the phase of the perturbation. The results partially support our hypothesis of a phase-independent functional balance response underlying the phase-dependent recruitment of different balance mechanisms at different points of the gait cycle. We conclude that the central nervous system recruits any available mechanism to push against the ground to maintain balance as fast as possible in response to a perturbation, but the different mechanisms do not have equal strength.

Sign in / Sign up

Export Citation Format

Share Document