Testing the Reliability and Validity of the XOS Motion Capture

System at Measuring Counter Movement Vertical Jump

Suzanne M. Konz and David Cottrill

Marshall University, Huntington WV, U.S.A.

Keywords: CMVJ, Measurement, Instrumentation, Reliability.

Abstract: The purpose of this study was to conduct simultaneous measurement of counter movement vertical jump

height using the XOS motion capture system and the Vertec system. Ten participants (body height: 170.17

cm ± 13.4, body weight: 79.76 kg ± 17.72) from the Marshall University student body comprised the testing

group. Participants were instructed on proper counter movement vertical jump technique. Five practice

jumps at 50% effort were conducted. Participants donned a compression suit with reflective markers. The

paired t-tests indicated that a difference existed in counter movement vertical jump height measured

between the Vertec and the XOS vertical jump was (p < 0.001), SEM of 1.4 with a .823 correlation and the

Vertec and the XOS center of gravity was also (p < 0.001), SEM of 1.42 with a correlation of .788. A

marked difference exists between the XOS SportMotion capture system’s methods of measuring counter

movement vertical jump height when compared to the Vertec measurement.

1 INTRODUCTION

Motion analysis systems are a widely used tool in

performance enhancement, biomechanical analysis,

and injury assessment. These systems provide users

with important information to guide the

improvement of function. The reliability and validity

of the information of these systems is vital.

Reliability is the degree to which an experiment,

test, or measuring procedure produces stable and

consistent results. Reliability for measurement

systems like these are concerned with concepts like

stability, reliability, and internal consistency

(Vincent, 2009). More importantly however, these

systems require validity to be able to be of value as

true measurement tool. The validity of a system tells

the user how well it measures what is supposedly

measures.

The XOS SportMotion system (Motion Reality,

Inc. Marietta, GA) is a relatively new technology

platform built upon the most modern advances in 3-

D Motion Capture and Analysis technology.

According to the company’s website, SportMotion is

the world's first 3-D motion capture system

specifically designed to help measure an athlete's

performance, aid in rehabilitation, assist in training

and become an effective teaching tool (Motion

Reality 2014). The technology of the XOS

SportMotion system is similar to that used to

produce movies and video games, but is customized

to specifically serve the functional and usability

needs for athletes. The system is marketed as a

convenient device to use to improve performance

within the strength and conditioning and team

specific areas. Several professional teams in the

MBL and NFL along with NCAA-I teams use this

system to improve athlete performance. It is not

used typically for quantitative research purposes.

A component of the XOS SportMotion system is

the measurement of counter movement vertical jump

height (CMVJH). This CMVJH data, normally

provided through physical measurement using a

Vertec (Vertec Sports Imports, Hilliard, OH)

measuring device, is typically generated through

tracking the subject’s center of mass (COM) (Isaacs,

1998). The difference between the resting height of

COM and the peak height during the jump is

presented as CMVJH. In addition to tracking COM

travel, certain systems, such as the XOS motion

capture system calculate CMVJH through

measurement of the time the subject is off the

ground. This method is employed by Jump Mat

systems, and has been found to be comparable to

Vertec and center of gravity (COG) tracking

methods (Isaacs 1998, Pond, Verducci et al. 2003,

Leard, Cirillo et al. 2007)

The Vertec CMVJ testing system is device

typically used by universities and high schools to

M. Konz S. and Cottrill D..

Testing the Reliability and Validity of the XOS Motion Capture System at Measuring Counter Movement Vertical Jump.

DOI: 10.5220/0005094500050010

In Proceedings of the 2nd International Congress on Sports Sciences Research and Technology Support (icSPORTS-2014), pages 5-10

ISBN: 978-989-758-057-4

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

test athlete vertical jump height. It is considered the

testing device of choice due to the low coast and

high reliability. The Vertec device requires an

athlete to maximally reach for the object while

jumping. The measuring device is widely used in

athletic testing due to its simplicity. The Vertec

(Vertec Sports Imports, Hilliard, OH) consists of a

series of colored plastic vanes that are placed 0.0127

m apart on a telescoping aluminum pole that can be

adjusted to the subject’s standing reach. The subject

performs a maximal jump and swats at the plastic

vanes at the peak of the jump. Vertical jump height

is measured as the vertical distance between the

standing reach and the highest vane displaced by the

subject’s hand at the peak of the jump. Jump height

assessed by the Vertec is determined by subtracting

the standing height or reach height by the maximum

jumping height or reach height using procedures

such as Sargent’s, Abalakov’s, and Starosta’s, jump

tests (Klavora, 2000; Starosta & Radzinska, 2001).

As such the reliability and validity of the XOS

SportMotion system is not known. To date, no

studies testing the reliability of the XOS system’s

measurement methods in comparison to the gold

standard Vertec measurement system exist

(

Hutchinson Issacs 1998, Petushek 2010, Pond 2003).

From a research perspective, the gold standard is

either 3-D motion analysis or force plate. The

question at hand is how reliable and valid is the

XOS SportMotion system. The purpose of this study

is to conduct simultaneous measurement of CMVJH

using the XOS motion capture system and the

Vertec system. The comparison of these results will

help determine the reliability and validity of the

XOS system in measuring jump height compared to

a verified measurement system.

2 METHODS

Prior to experimental testing, project approval was

obtained from the Marshall University Institutional

Review Board. Ten participants (body height:

170.17 cm ± 13.4, body mass: 79.76 kg ± 17.72)

from the Marshall University student body

comprised the testing group. Participants included

four male (body height: 177.80 cm ± 9.51, body

mass: 81.13 kg ± 8.45) and six females (body

height: 165.09 m ± 14.67, body mass: 78.85 kg ±

23.77). All subjects signed informed consent and

were able to withdraw at any time during the course

of the study.

Clothing/

Strap

arkers

Location

1 Shirt 12 2 markers top of shoulders

3 markers across top of back

2 markers in center of back

1 marker on sternum

2 markers on side of each upper arm

1 Belt 4 2 markers side of waist

2 markers back of waist

1 Cap 4 2 markers on top of head

2 markers in front

2 Wrist

Straps

2/wrist 1 marker on outside of wrist

1 marker on inside of wrist

2 Knee

Straps

2/knee 1 maker centered below knee on leg

1 marker on outside of shin

2 Shoe

Covers

4/foot 2 markers on top of shoe

1 maker at center of heel

1 maker centered on outside of foot

Figure 1: “Marker placement”.

The XOS Sport Motion system (Motion Reality,

Inc. Marietta, GA) is an infrared tracking system

that provides instant three-dimensional motion

feedback to assist in the training and performance

evaluation of athletes for all levels. The XOS

system housed in this laboratory utilizes 24 cameras.

Each XOS Sport Motion system contains three

options for specific data collection: golf, baseball,

and free play. The option selected determines the

size of the calibration space and the number of

cameras used. The golf option uses the smallest

space and only eight of the 24 cameras. The free

play option uses the largest space and all 24 camers.

The free play option allows the user to be creative

with the data collected. Items that can be tested

include vertical jump, broad jump, and tracking of

weight lifting technique. This study utilized the free

play option. Multiple requests for further

information on the process by which the system

processess and calculates were not met due to the

proprietary nature of this information.

The XOS Sport Motion system was calibrated

each testing day according to the systems required

means. This is a two phase process. The first phase

begins with placing the reflective wand on three

specific locations spaced 3 meters apart followed

byplacing the wand in locations around the oustide

of the free play area in a vertical direction. The

system notifies the user when all 24 of the cameras

have recognized the wand and the location

plaements. The second phase is the typical sweeping

of the inside of the calibrated space. The system then

combines these to actions to calibrate the space. The

system is ready for data collection with a successful

calibration.

icSPORTS2014-InternationalCongressonSportSciencesResearchandTechnologySupport

Users of the XOS Sport Motion don a compression

suit with 36 reflective markers located at specific

locations (See Figure 1). Four additional markers are

attached to the suit as part of the global reference

component. Two global markers are placed on the

anterior chest at the shoulder joint area. The other

two global markers are placed in the general

location of the hip, usually over the greater

trochanter. This makes a total of 40 reflective

markers associated with the compression suit. The

additional four markers are removed once an avatar

is generated. These markers allow the XOS system

to generate an avatar model that is displayed to

allow the athlete to view the skill for feedback. The

avatar comes in only two versions: a male and

female avatar. The avatar adjusts its look based upon

the distribution of the markers in the known pattern

for the individual

The 28 reflective markers used for data

collection are not placed on the joint axis as required

by most infrared tracking systems. Rather, the

markers are placed in unique placements to fit a

specific pattern (See Figure 1). The interesting

component here is that the locations per the user

guide manual are very generalized locations.

However, during training the representatives used

anatomical reference points for the location of

several of the markers.

Participants for this study donned the

compression suit. 36 permanent and 4 temporary

global markers were placed according to company

standards. Participants were instructed on proper

CMVJ (counter movement vertical jump) technique

and use of the Vertec (Vertec Sports Imports,

Hilliard, OH) during CMVJ testing. Reach height

for the Vertec was established using the following

body position: erect stance, both feet together and

ﬂat on the ground, both arms fully extended

overhead, and the head and eyes level (Issacs 1998).

Instructions on the CMJ technique were then

provided. This technique required subjects to start in

an upright position with the feet parallel to each

other and hip to shoulder width apart. The subject

then performed a quick countermovement drop into

a quarter-squat position by ﬂexing the hips and

knees into a semi-squat position while swinging

their arms back to prepare for the jump. After

reaching their preferred depth of descent, subjects

explosively extended at the knees and hips, and

plantar ﬂexed at the ankles in an effort to attain a

maximal jump height. During the concentric and

ﬂight phases of the jumps, subjects were required to

maintain a level head position (i. e., not looking

Figure 2: “T-position”.

upward at the Vertec vanes) while reaching upward

with both hands simultaneously (Issacs 1998).The

arms swing forward above their head as they jump

straight up into the air, landing on both feet at the

same time (Harman 1990). Arm swing has been

shown to inﬂuence vertical jump height (Lees 2004)

and performance biomechanics (Lees 2004).

Five practice jumps at 50% effort were

conducted to ensure understanding of appropriate

technique. A rest period of at least 60 seconds

between each jump occurred during familiarization

to provide feedback on improving the participant’s

technique along with recovery. After familiarization

was complete, participants left the room to allow for

a noise elimination procedure which is required by

the XOS system. This part of the avatar generation

process required by the system. Upon completion of

the noise elimination, the participants re-entered the

room and took their place within the calibrated

space. The system began the process of generating

an avatar model for each participant at this time.

This was accomplished by having the participant

stand within the calibrated space in a “t-position” as

the system went through the process of recognizing

the reflective marker pattern. The t-position finds the

subject standing in an erect posture with the feet

approximately shoulder width apart while the

shoulders are abducted to approximately ninety

degrees (See Figure 2). The participant’s avatar is

generated after the system recognizes the reflective

markers being in the correct configuration and

locations.

With the avatar generated, participants again

entered the calibrated space and conducted three

CMVJ trials separated by 60 seconds of rest. During

these trials, jump height was measured

TestingtheReliabilityandValidityoftheXOSMotionCaptureSystematMeasuringCounterMovementVerticalJump

simultaneously by the XOS system and the Vertec.

Vertec data was collected by the same researcher

who provided the instruction on CMVJ technique.

The XOS data measured the calculated center of

gravity travelled and vertical jump height through

proprietary software. Data was analyzed using SPSS

(IBM, Armonk, New York). Descriptive statistics,

paired t-tests, and intraclass correlation coefficients

(ICC 1,3 and ICC 2,3) analysis were completed.

Significance was set at the 0.05 level. ICC 1,3 was

run to investigate the reliability and validity for each

type of CMVJ test. An ICC 2,3 determined the

reliability of the XOS Sport Motion CMVJ testing

against the Vertec.

3 RESULTS

Descriptive statistics are presented in Table 1. The

paired t-tests indicated that a difference existed in

CMVJ height measured. The significance for the

comparison between CMVJ height measured

between the Vertec and the XOS VJ (XOS vertical

jump) was (p < 0.001), SEM(standard error of the

mean) of 1.4 with a .823 correlation The

significance for the comparison between CMVJ

height measured between the Vertec and the XOS

COG (XOS center of gravity) was also (p < 0.001),

SEM of 1.42 with a correlation of .788.

The vertical jump height measured with the

Vertec ranged from 31.75 cm to 82.55 cm. The

vertical jump height measured with the XOS VJ

ranged from 23.68 cm to 61.47 cm. The reliability

(ICC 1,3) of the Vertec measures was 0.97. The

(Standard error of Measurement) for the Vertec

measures was 0.0004. A MCD (minimal clinical

difference) for the Vertec was 3.58. The reliability

(ICC 1,3) of XOS VJ measures was 0.936. The SE

for the XOS VJ measures was 0.016 with an MCD

of 2.69. The reliability (ICC 2,3) for the Vertec and

the XOS VJ was .871.

The vertical jump height measured with the XOS

COG ranged from 30.734 cm to 62.23 cm. The

reliability (ICC 1,3) of the Vertec measures again

was 0.97. Again, the SE

for the Vertec measures

was 0.0004. An MCD for the Vertec was 0.005. The

reliability (ICC 1,3) of XOS COG measures was

0.945. The SE

for the XOS COG measures was

2.46 cm. And, a MCD calculated at 3.54. The

reliability (ICC 2,3) for the Vertec and the XOS

COG was .833.

Table 1: Descriptive Statistics.

Device Group Mean ± Std. Dev. (cm)

Vertec

All 49.66 ± 12.56

M 59.16 ± 13.41

F 43.32 ± 6.82

XOS VJ

All 39.38 ± 13.46

M 53.41 ± 8.83

F 30.03 ± 5.53

XOS COG

All 43.46 ± 9.42

M 52.87 ± 5.65

F 37.24 ± 5.23

4 DISCUSSION

All three means of measurement demonstrated

adequate individual reliability. This means that the

each of the systems measured CMVJH consistently.

However, the validity of the XOS system’s

measurements did not prove as great as the Vertec.

An interesting situation was noted with two of our

subjects that demonstrated part of the problem with

the internal consistency with the XOS system. Two

subjects (subject 5 and 9) had Vertec measurements

of 82.55 cm for their CMVJ. XOS SportMotion

calculated the XOS VJ at 66.55 cm and 53.34 cm for

subject 5’s CMVJ heights. Subject 9’s CMVJ height

at 82.55 cm was calculated at 54.61 cm. These

differences show that there is a lack of consistency

within the calculation of XOS VJ height.

XOS SportMotion system has two definitions

attached to the label "COG". One is used to calculate

the COG path (actual and floor projected) and the

other is used for the calculation of the vertical and

horizontal jump functions. In the vertical jump

function, the 3D location designated as the COG is

actually approximated to the origin of the waist

body in the skeleton (See Figure 1). During the

scaling process, the system optimizes this location

based on the placement of the markers, for both

capture and scaling, identified during said scaling

process. The vertical distance measurement is the

difference between the take-off height and peak

height of this COG location; where the take-off

frame is calculated as the frame where both feet

have been deemed to have left the floor plane. The

feet are calculated to have left the floor when both

heels are more than 4 inches above the floor plane.

The heel is approximated as the points located 3

inches below each ankle. Landing occurs at the

frame where at least one of the heel locations is back

icSPORTS2014-InternationalCongressonSportSciencesResearchandTechnologySupport

within 4 inches of the floor plane. This method of

calculation does not take in to account that most

individuals will land on the forefoot to provide a

triple absorption of force through ankles, knees, and

hips (Motion Reality 2014). With information

provided by Motion Reality, Inc, the XOS system

software appears to calculate jump height by using

total time the subject spends off the ground. These

XOS COG data seems to calculated with the

following equation:











∗



 ,

where t represents time off the ground and G the

gravitational constant to confirm or refute this

assumption (Isaacs 1998, Pond, Verducci et al.

2003, Leard, Cirillo et al. 2007). However, we could

not get this confirmed by the company.

The company’s marker placement may be a

source of error when it comes to the reliability and

validity of the XOS Sport Motion. The traditional

infrared 3-D motion analysis system requires the

placement of reflective markers at the joint centers

to assist in determining segment lengths, kinematics,

and kinetics. The XOS system is looking for specific

patterns and not locations to develop the avatar. The

company instructions on marker location are part of

the issue.

The shirt has 12 markers placed on it. Two

markers are placed on the top of shoulders, three

markers across top of back, two markers in center of

back, one marker on sternum, and two markers on

side of each upper arm. The instructions don’t give

clear expectations of this placement. The shoulder

makers are placed on the AC joint. The three

markers on the top of the back are equally

distributed across the back. The center markers on

the back again are distributed equal at the mid-back

level of the participant. The marker on the sternum

is located on the upper portion. And the two markers

on the side of the arm are placed at the elbow and at

a location that is 1/3 down the upper arm from the

shoulder marker. Unless you had knowledge of the

location from training by company representatives

you would not know the locations the system

expects in order to recognize and generate the

avatar.

The belt worn at the waist requires four markers:

two on the side and two on the posterior side. The

system expects these markers at the ASIS and L4-L5

location. The cap also has 4 markers. 2 are located at

the front and 2 on top. An issue occurs with

overweight individuals at the ASIS markers. The

markers on the belt rotate downward toward the

floor due to the material of the belt. The altered

positions make it difficult for the cameras to see the

markers. The cap has four markers as well. The

instructions list 2 at the front and 2 on the top of the

head. However, the system wants one marker at

either temple, one on top of the head, and one at the

back of the head. This posterior head marker

becomes an issue with females having long hair.

The wrist, knees, and feet straps the company

uses also present challenges for the system to

recognize. Each of the wrist and knee straps has two

markers for each of the extremities. The wrist strap

help the system understand pronation and supination

of the forearm. In order to accomplish this, the

system needs to see an offset of the markers at the

wrist. However, the instructions provided by the

company lists that one marker be attached on outside

of wrist and one marker on inside of wrist. The

system does have a hard time determining which

marker is on the outside and which is on the inside

of the wrist. As a result, the avatar does not always

generate a correct model or the model will have a

“twitch” in the hand and wrist area. The feet require

the subject to wear shoes and covers are placed over

the participant’s shoes. Each shoe cover has four

markers. The four markers are instructed to be

located at on top of shoe, at center of heel, and

centered on outside of foot. In reality, the two

markers need to be located on the great toe and 5

digit, one marker is located at the center of the heel,

and one marker is located on the 5

metatarsal. An

issue here is the size of the shoe covers does not

allow for the larger feet of many athletes. This

makes it difficult for appropriate marker locations to

be provided.

The system introduces error into the calculations

provided to users in a couple ways. The calculation

of the jump height provides much of the error.

Marker placement is also a source of error. Both of

these lend to decrease reliability and validity on the

XOS Sport Motion system.

5 CONCLUSIONS

Based on initial data analysis, there is a marked

difference between the XOS SportMotion capture

system’s methods of measuring CMVJ height when

compared to Vertec measurement. XOS

SportMotion does provide a reliable means of

measuring CMVJ; however, the measurements

provided are not at the same level of validity as the

Vertec system. Individuals using the XOS

TestingtheReliabilityandValidityoftheXOSMotionCaptureSystematMeasuringCounterMovementVerticalJump

SportMotion system need to keep this in mind when

using this particular component to evaluate athlete

performance. Interpretation of these results confines

generalization to recreationally active college-aged

students. Future studies should test other suitable

populations such as the athletes.

REFERENCES

Feltner, M.E. Bishop, E.J., and Perez, C.M. (2004).

Segmental and kinetic contributions in vertical jumps

performed with and without an arm swing. Res. Q.

Exerc. Sport 75:216–230. 2004.

Harman, E.A., Frykman, P.N., Roesnstein, R.M. (1990)

The effects of arms and countermovement on vertical

jumping. Med. Sci. Sports Exerc. 22: 825–833.

Hutchinson A.T. and Stone, A.L. (2009) Validity of

Alternative Field System for Measuring Vertical Jump

Height. JEPonline. Vol. 12(3) June 2009.

Isaacs, L. D. (1998). "Comparison of the vertec and Just

Jump Systems for measuring height of vertical jump

by young children." Percept Mot Skills 86(2): 659-

663.

Lees, A., Vanrenterghem,J. and Clercq, D.D. (2004)

Understanding how an arm swing enhances

performance in the vertical jump. J. Biomech. 37:

1929.

Klavora, P. (2000) Vertical-jump: a critical review.

Strength and Conditioning Journal, 22(5), 70-75.

Leard, J. S., M. A. Cirillo, E. Katsnelson, D. A. Kimiatek,

T. W. Miller, K. Trebincevic and J. C. Garbalosa

(2007). "Validity of two alternative systems for

measuring vertical jump height." J Strength Cond Res

21(4): 1296-1299.

Motion Reality, I. (2014). Personal communication. S.

Konz.

Petushek, E.J., VanderZanden, T., Wurm, B., and Ebben,

W. P. (2010) Comparison of Jump Height Values

Derived from a Force Platform and Vertec.

International Symposium on Biomechanics in Sports:

Conference Pr;, Vol. 28.

Pond, J., F. Verducci and B. Scheilhauf (2003). Reliability

and Validity of the Just Jump System. M.S., San

Francisco State University.

Starosta, W. and Radzinska, M. (2001) Comparison of

jumping ability measured by different methods in

sportsmen. Biology of Sport 18, 245-251.

Vincent, W.J. and J. Weir. Statistics in Kinesiology 4th ed.

2012. Human Kinetics. Champaign, IL

icSPORTS2014-InternationalCongressonSportSciencesResearchandTechnologySupport