Skip to main content

Table 1 Summary of the main validation studies analyzing motion capture accuracy of the Kinect (v1 and v2) and of wearable IMUs

From: Kinect and wearable inertial sensors for motor rehabilitation programs at home: state of the art and an experimental comparison

Refs.Sensor(s)ParticipantsExercisesParametersApplicationsPerformance
[49]Kinect v11H6 primary movementsJoints positions, bone lengthsAt-home elderly rehabilitationRMSE \(< 10cm\)
[50]Kinect v120HReaching, standing balanceJoints positionsPostural controlICC \(< 0.16\)
CV \(<11.6\%\)
[51]Kinect v148H, 2 sessions4 primary movementsJoints anglesFunct. assessmentRMSE \(<9\%\)
[52]Kinect v110H + 9PD12 primary and functional movementsJoints positions and angles, timingsPD assessmentLoA \(<10\%\)
ICC \(>0.9\)
[53]Kinect v120HStanding swayMarker positionsExergamingSDev \(<30\%\)
[54]Kinect v115HSquatLower limb joints anglesKinect evaluationRMSE \(<5^\circ\)
ICC \(>0.9\)
[36]Kinect v230H, 2 sessionsStanding balanceTrunk and pelvis anglesBalance and postural controlr \(>0.75\)
ICC \(>0.7\)
[42]Kinect v230H, 2 sessionsWalkingLower limbs joints anglesGait analysis\(0.4<\) r \(< 0.75\)
[39]Kinect v220H5 primary shoulder movementsUpper limbs joints anglesClinic and home rehabilitationRMSE = \(3.9^\circ\)
[43]Kinect v220HWalkingStance and step time, step length, time, velocityGait analysisLoA \(<11\%\)
ICC \(>0.9\)
[41]Kinect v230HStanding and sitting exercisesUpper body joint positionsRehabilitationRMSE \(<5\%\)
r \(>0.9\)
[55]4 IMU (custom) on thighs, shanks3HWalkingHip and knee joint anglesGait analysisRMSE \(<8.7^\circ\)
r \(>0.7\)
[6]2 IMU (Xsens MTx) on pelvis, thigh20HWalkingHip joint angleGait analysisLoA \(= 1.6^\circ\)
[56]8 IMU (Xsens MTx) on thorax, lower back, thighs, shanks, feet9H (children)Static posturesHip and knee joint anglesKnee amputees and cerebral palsyRMSE \(=1.8^\circ\)
[57]6 IMU (not specified) on pelvis, thigh, shank9H + 1INJRunning and outdoor training routinesActivity classification, hip and knee joints anglesAthlete training, injury prevention98%
[58]3 IMU (Shimmer) on thigh, shank, foot58REHPrimary movementsExercise recognitionClinic and home rehabilitation83%
[8]5 IMU (Riablo) on trunk, thigh, shank22H, 2 sessions4 knee exercisesHip and knee joint anglesClinic and home rehabilitationRMSE \(=3.1^\circ\)
[59]2 IMU (Opal) on lower back, upper arm6HADL routines,
walking
Trunk and upper limb joints anglesIMU evaluationRMSE \(<5.5^\circ\)
[60]6 IMU (Xsens MTx) on thigh, shank, foot1AMPWalkingKnee angleGait analysisRMSE \(<3^\circ\)
[61]7 IMU (Opal) on trunk, upper arm, forearm, hand8H2 swimming stylesUpper limb joints anglesswimming analysisRMSE \(<7^\circ\)
[62]3 IMU (Opal) on waist, thigh, shank, 1 resistive strip14HSit-to-stand, walkingExercise timings, stride lengthClinic and home rehabilitationr \(>0.7\)
ICC \(>0.95\)
[63]17 IMU (Synertial) full body20H5m timed up and goFull body jointsClinic evaluationRMSE \(<13.6^\circ\)
[64]2 IMU (APDM)
thigh and shank
18REHKnee ROM, walkingKnee angleClinic rehabilitation
[9]17 IMU (Xsens MVN) full body12HFunctional movements, work actionsFull body modelOccupational biomechanicsRMSE \(<5^\circ\)
[65]3 IMU (I2M MT) on pelvis, trunk, upper arm6H lab + 10H workplaceWork actionsTrunk orientation, shoulder joint anglesOccupational biomechanicsRMSE \(<12.1^\circ\)
[47]Kinect v1, Kinect v220H (v1) + 20H (v2)15 static posesJoints positionsPosture evaluationRMSE(v1) \(=76mm\)
RMSE(v2) \(=87mm\)
[48]Kinect v1, Kinect v210H, 3 camera anglesSitting and standing posturesJoints positionsMotion captureRMSE \(<100mm\)
[33]Kinect v1, Kinect v213HROM and static stretching exercisesJoints positionsRehabilitationRMSE \(=15mm\)(wrist)
RMSE \(<10mm\)(other)
[66]Kinect v1 + 2IMU (InertiaCube3) on upper limb1HHand-to-mouthUpper limb positionMotion capture
[67]Kinect v1 + 1 x-IMU1HKnee flexion/extensionKnee angleMotion captureRMSE \(<14.2^\circ\)
[68]Kinect v1 Vs Kinovea (MBS) Vs Hillcrest (IMU)2HWalkingHip and knee anglesGait rehabilitationRMSE \(=3^\circ\) (IMU)
RMSE \(=10^\circ\) (Kin.)
  1. Participants are labeled as healthy (H), Parkinsons (PD), injured (INJ), in rehabilitation (REH) and amputees (AMP). All experiments were conducted in lab unless otherwise stated. Reported performance metrics include Root Mean Square Error (RMSE), Interclass Correlation Coefficient (ICC), Coefficient of Variation (CV), Limits of Agreement (LoA), Standard Deviation (SDev) and Pearson’s correlation coefficient (r)