MocapApi Calc type data in detail

The Calc data in MocapApi mainly contains six types of data, this article details the definition,
format, and coordinate system information of this data.

Introduction

Calc Data can be exported into a .CSV or broadcasted from Axis Studio. 

Overview of Calc data items

The Calc data item is divided into 4 segments, and the segments are connected by:

• The first paragraph is the name of the body part, such as Hips, SpineBottom, …, LeftHandPinkyTip, etc. 59 kinds, please refer to the “First paragraph (body part) description” below for details.
• The second segment is feature information, including three types of sensor, joint joint, and bone bone.
• The third section is physical quantity information, including attitude quaternion Quat, angular velocity Gyro, acceleration Acce, velocity Velo, and position Posi. Also includes the packet loss marker Lost.
• The fourth paragraph is the specific component value, including x, y, z components, of which the quaternion has one more w component.

In the Calc data, a total of 6 different data are combined from the second and third paragraphs above for output, including:

• Sensor attitude quaternion Sensor-Quat
• Sensor Angular Velocity Sensor-Gyro
• Sensor acceleration Sensor-Acce
• Joint Velocity Joint-Velo
• Joint position Joint-Posi
• Bone Pose Quaternion Bone-Quat.

Body Part Description

• In the data output by MocapApi, the human body is divided into 59 parts, of which the hip node is the root node, and each subsequent node is connected to a parent node, the specific order is shown in the following table. This connection is used for the Joint-Velo and Joint-Posi data in the Calc data. The definition of each node
• Joint is the connection between this node and its parent node. For example, the parent node of “right forearm” is “right forearm”, and its connection is the elbow joint of the right arm, so the Position and Velocity of “right forearm” both refer to the Position and Velocity of the elbow joint of the right arm.

Section Name	Logotype	serial number	Parent Node
Buttocks	Hips	0	Root Node
Right thigh	RightUpLeg	1	0
Right Calf	RightLeg	2	1
Right foot	Rightfoot	3	2
Left thigh	LeftUpLeg	4	0
Left calf	Leftleg	5	4
Left foot	LeftFoot	6	5
Lower Part of the Spine	Spine	7	0
Middle Spine section	Spine 1	8	7
Upper Spine section	Spine 2	9	8
Lower Neck section	Neck	10	9
Upper Neck section	Neck 1	11	10
Head	Head	12	11
Right Shoulder	RightShoulder	13	8
Right Arm	RightArm	14	13
Right Forearm	RightForeArm	15	14
Right Hand	RightHand	16	15
Right thumb finger	RightHandThumb1	17	16
Right thumb in the middle finger	RighthandThumb2	18	17
Right Thumb tip	RighthandThumb2	19	18
Right index metacarpal	RightInHandIndex	20	16
Right index finger root	RightHandIndex1	21	20
Middle finger of the right index finger	RightHandIndex2	22	21
Right index fingertip	RightHandIndex3	23	22
Right middle metacarpal	RightInHandMiddle	24	16
Right middle finger to the root	RightHandMiddle1	25	24
Right middle finger middle	RightHandMiddle2	26	25
Right middle fingertip	RightHandMiddle3	27	26
Right ring metacarpal	RightInHandRing	28	16
Right ring finger refers to the root	RightHandRing1	29	28
Right ring finger in the middle	RightHandRing2	30	29
Right ring fingertip	RightHandRing3	31	30
Right little finger metacarpal	RightInHandPinky	32	16
Right pinky finger root	RightHandPinky1	33	32
Right pinky finger in the middle	RightHandPinky2	34	33
Right pinky fingertip	RightHandPinky3	35	34
Left shoulder	LeftShoulder	36	8
Left upper arm	LeftArm	37	36
Left forearm	LeftForeArm	38	37
left hand	LeftHand	39	38
Left thumb finger root	LeftHandThumb1	40	39
Left thumb in the middle finger	LeftHandThumb2	41	40
Left thumb tip	LeftHandThumb3	42	41
Left index metacarpal bone	LeftInHandIndex	43	39
Left index finger root	LeftHandIndex1	23	43
Middle finger of the left index finger	LeftHandIndex2	23	44
Tip of the left index finger	LeftHandIndex3	23	45
Left middle metacarpal	LeftInHandMiddle	23	39
The left middle finger refers to the root	LeftHandMiddle1	23	47
The left middle finger is fingered in the middle	LeftHandMiddle2	23	48
Left middle fingertip	LeftHandMiddle3	23	49
Left ring metacarpal	LeftInHandRing	23	39
The left ring finger refers to the root	LeftHandRing1	23	51
Left ring finger in the middle	LeftHandRing2	23	52
Left ring fingertip	LeftHandRing3	23	53
Left little finger metacarpal bone	LeftInHandPinky	23	39
Left little finger finger root	LeftHandPinky1	23	55
Left little finger in the middle	LeftHandPinky2	23	56
Left little fingertip	LeftHandPinky3	58	57

Feature Information

• Sensor Sensor: In inertial motion capture, some body parts are not bound with corresponding sensors, and
  the “sensor data” of these body parts are back-calculated from body skeleton data. At this stage, MocapApi
  only supports Perception Neuron Studio sensors.
• Joint Joint: explained in the previous section “First paragraph (body part) description”
• Bone Bone: that represents the body part

Global coordinate system

All physical quantities in the Calc file output by MocapApi take the upper left front coordinate system of the human body calibration pose as the global coordinate system. If there is no human body calibration (such as Axis Lab software), use the North Sky East coordinate system as the global coordinate system

Sensor body coordinate system

Perception Neuron Studio sensor, its sensor after the next computer AHRS calculation, the host computer received the output data of the ontological coordinate system as shown in the following figure.

The Perception Neuron 3 sensor, after the sensor is calculated by the lower computer AHRS, the ontological coordinate system of the output data received by the host computer is shown in the following figure.

The coordinate system of the human body

The definition of the human node coordinate system follows the definition of the BVH format, and the coordinate system of each node output data is the upper left coordinate system in the case of TPose, as shown in the following

Position zero

The Calc output data contains the Node’s Pose data (Joint-Posi), whose zero point is defined as the intersection of the red and blue lines on the ground in axis Studio software, and it is also the projection of the location of the Hips node on the ground after Avatar has completed the human posture calibration. As shown in the following

Rotation Zero Point:

The Calc output data contains the rotation data (Bone-Quat) of each bone of the body, the zero point of
which is defined as the angle of each node when the human body faces the Z axis in the case of TPose For the specific definition of T-Pose, please refer to the BVH format header file description of the Perception Neuron Studio skeleton

Physical Quantities

• Quaternion Quart: Expresses the amount of rotation of the sensor and bone in a standard quaternion format. The rotation is consistent, but there are two kinds of data outputs because of the different coordinate systems.
• Angular velocity Gyro: The unit is the radian system, that is, rad/s
• Acceleration Acce: The unit is g
• Velocity Velo: TSUs m/s
• Location Posi: INTERNATIONAL STANDARD unit m

Calc Data is illustrated with specific examples.

• The data item name (table header) is generally divided into 4 segments, and the segments are connected with - as explained above.
• Frame sequence number, marked from 0.
• The packet loss is marked Loss, the value 1 represents the packet loss, and the data value is obtained by the algorithm. The value 0 is normal, and the data value is the sensor measurement.
• An example is shown in the following table.

Data item header description	illustrate
Frame-No	Frame number
Hips-Sensor-Lost	HIPS partial sensor packet loss marker
Hips-Sensor-Quat-x	HIPS part sensor attitude quaternion x component
Hips-Sensor-Quat-y	HIPS partial sensor attitude quaternion y component
Hips-Sensor-Quat-z	HIPS partial sensor attitude quaternion z-component
Hips-Sensor-Quat-w	HIPS partial sensor attitude quaternion w component
Hips-Sensor-Gyro-x	HIPS partial sensor angular velocity x component
Hips-Sensor-Gyro-y	Hips partial sensor angular velocity y component
Hips-Sensor-Gyro-z	Hips partial sensor angular velocity z-component
Hips-Sensor-Acce-x	HIPS partial sensor acceleration x component
Hips-Sensor-Acce-y	HIPS partial sensor acceleration y component
Hips-Sensor-Acce-z	Hips partial sensor acceleration z-component
Hips-Joint-Velo-x	Joint velocity x component of the HIPS section with parent node
Hips-Joint-Velo-y	Joint velocity y component of the HIPS fraction with the parent node
Hips-Joint-Velo-z	Joint velocity z component of the HIPS section with the parent node
Hips-Joint-Posi-x	Joint position x component of the hips section with parent node
Hips-Joint-Posi-y	The joint position y component of the HIPS fraction with the parent node
Hips-Joint-Posi-z	The joint position z component of the HIPS section with the parent node
Hips-Bone-Quat-x	Hips partial skeletal pose quaternion x component
Hips-Bone-Quat-y	Hips partial skeletal pose quaternion y component
Hips-Bone-Quat-z	Hips partial skeletal pose quaternionion, z-component
Hips-Bone-Quat-w	Hips partial skeletal pose quaternion w component
SpineBottom -Sensor-Lost	SpineBottom Part of the sensor drop marker
SpineBottom -Sensor-Quat-x	SpineBottom part sensor attitude quaternion x component
SpineBottom -Sensor-Quat-y	SpineBottom part of the sensor attitude quaternion y component
SpineBottom -Sensor-Quat-z	SpineBottom part of the sensor attitude quaternion z-component
SpineBottom -Sensor-Quat-w	SpineBottom part of the sensor attitude quaternion w component
-----------	-----------
LeftHandPinky1-Bone-Quat-x	LeftHandPinky1 part skeletal pose quaternion x component
LeftHandPinky1-Bone-Quat-y	LeftHandPinky1 partial skeletal pose quaternion y component
LeftHandPinky1-Bone-Quat-z	LeftHandPinky1 part skeletal pose quaternion z-component
LeftHandPinky1-Bone-Quat-w	LeftHandPinky1 part of the quaternion w component of skeletal pose

Output file format

Axis Studio export for this file type can be found under the export window. Once selected, users can only choose the Range of frames to export. 

• The file extension is csv, and data items are separated by commas.
• The total number of frames for row 1 of the file.
• Line 2 of the file is a description of the data item header. Use the header instructions from the table above. The header order is: the first column is Frame-No, and then output in the order of the serial number of 59 body parts, each body part has 21 data, and output in the order according to the above header description. So the total number of columns in the data is: each body part has 21 quantities [(Lost marker) X1+ (Sensor Quaternion) X4+ (Sensor Gyro) X3+ (Sensor Acce) X3+ (Joint Velocity) X3+ (Joint Position) X3+ (Bone Quaternion) X4], a total of 59 body parts, including the frame number, so each 1 line is a total of 59* 21+1=1240 items of data, that is, 1240 columns of data.
• The file starts from line 3 for the data part, and the data is 1 line every 1 frame. The data is separated by commas, a total of 1240 columns.
• An example is as follows. The data file named "Data" has a total frame count of 998 frames.

data.csv

998

Frame-No,Hips-Sensor-Lost,Hips-Sensor-Quat-x,Hips-Sensor-Quat-y, ..., LeftHandPinky3-Bone-Quat-w

0, 0, 0.0, 0.0, 0.0, 1.0, -0.0036, -0.4065, 0.6567, -0.0036, -0.4065, 0.6567, ... , 0.0,0.0,0.0,1.0

1, 0, 0.0, 0.0, 0.0, 1.0, -0.0038, -0.4069, 0.6567, -0.0033, -0.4087, 0.6560, ... , 0.0,0.0,0.0,1.0

2, 1, 0.0, 0.0, 0.0, 1.0, -0.0138, -0.4169, 0.6267, -0.0133, -0.4387, 0.6561, ... , 0.0,0.0,0.0,1.0

............

........

997, 0, 0.0, 0.0, 0.0, 1.0, -0.0036, -0.4065, 0.6567, -0.0036, -0.4065, 0.6567, ... , 0.0,0.0,0.0,1.0