HP Reverb G2

Demo and values

VR headset from HP with four 640×480 grayscale tracking cameras and an IMU at about 1 kHz. All four are solved from the Monado SLAM calibration data, in three pair recordings stitched together.

Recovered by MVIS from the Monado calibration data.

Reprojection median0.13 px

cam0–cam1 baseline108.5 mm

cam0 td−1.25 ms

cam0 lever arm53.8 mm

Reprojection is how far, in pixels, the detected corners lie from where the solve puts them. td is the camera–IMU time offset, and the lever arm is the camera–IMU position offset.

One recording shown, with no independent reference on this page. These values show what MVIS solved, not how close they are to the truth.

Calibration process

The path of the rig around the board, as MVIS solved it. To explore it:

• Press Play or drag the bar to follow the path.
• Use the view buttons to change the view.
• Pick a camera card to put that camera on the frustum, the pyramid that shows its view.

Each card draws the board from the solved pose, with the corners MVIS detected. We publish no camera images.

Solved rig

The solved rig in 3D, to scale. Tap a sensor to see its values. In full screen you can rotate and zoom. Grid squares are 20 mm.

3DFrontTopSide+−

To scale, from above and the front

From above

From the front

Camera	Lever x, y, z (mm)	Distance (mm)	log(R) (rad)	td (ms)	Field of view
cam0	−51.9, +13.3, +5.1	53.8	[−0.345, +0.075, +1.538]	−1.25	118° × 94°
cam1	+56.6, +14.1, +5.4	58.6	[−0.074, +0.345, +1.594]	−1.25	119° × 95°
cam2	−75.7, +15.0, −30.3	82.9	[+0.954, −1.330, −1.068]	−1.29	118° × 94°
cam3	+80.8, +16.4, −29.8	87.7	[−1.383, +0.978, −1.377]	−1.35	118° × 94°

Solved by MVIS. Lever arms (camera–IMU position offsets) are in the IMU frame. Rotation is log(R), camera in the IMU frame.

Specifications

From the dataset
Cameras	4 × 640×480 grayscale at 30 Hz
IMU	About 1 kHz
Measured in these recordings
Calibrated field of view	cam0 118° × 94° · cam1 119° × 95° · cam2 118° × 94° · cam3 118° × 94°
Not published
Open	Sensor part numbers

Dataset

Recordings	Monado MGC round 1: imucalib01, 02 and 13, one for each camera pair
Source	Monado SLAM Datasets, Collabora
Board	AprilGrid 6×6, tag 30 mm, spacing 0.3, t36h11
Licence	CC BY 4.0
Hosted copy	Calibench, converted/monado_mgc/ (access by automatic approval)
Configs and results	Calibench community, hp-reverb-g2/ (access by automatic approval)
Data	Monado SLAM Datasets

AprilGrid 6 × 6, 225 × 225 mm. Standard pattern, to scale

Detailed results

Lens Calibrated field of view

Calibrated field of view, horizontal × vertical. cam0: 118° × 94°, corners reach 65° from the axis · cam1: 119° × 95°, corners reach 69° from the axis · cam2: 118° × 94°, corners reach 67° from the axis · cam3: 118° × 94°, corners reach 64° from the axis

cam0 cam1 cam2 cam3

We read these from the calibrated lens fit at the image edges, through the principal point (where the optical axis meets the image). ~ marks an axis past where the corners reached, so the fit is extrapolated there. Past the fit means an edge lies beyond where the fit holds, so we give no value.

IMU Gyro–accel time offset t_ga +2.54 ms

Each bar shows a scale as its deviation from 1. IMU model: calyx. t_ga, the time offset between the gyroscope and accelerometer channels: +2.54 ± 0.20 ms.

Gyro scale x+0.18 %

Gyro scale y+0.18 %

Gyro scale z+0.04 %

Accel scale x+0.78 %

Accel scale y+1.21 %

Accel scale z+0.97 %

−8 %0+8 %

Timing cam0 td −1.25 ms

The time offsets, to scale. td is the camera–IMU time offset, with t_imu = t_cam + td.

Full results Every solved parameter

Rotation is log(R) in rad, camera in the IMU frame.

Parameter	MVIS
cam0
fx px	269.92
fy px	270.20
cx px	322.51
cy px	229.16
k1	0.08280
k2	−0.04144
k3	0.22653
k4	−0.12781
Lever x mm	−51.85
Lever y mm	+13.32
Lever z mm	+5.13
log(R) rad	[−0.3452, +0.0746, +1.5384]
td ms	−1.250
Reprojection px	0.284
cam1
fx px	268.79
fy px	268.74
cx px	323.76
cy px	236.97
k1	0.07842
k2	−0.03462
k3	0.21848
k4	−0.12403
Lever x mm	+56.63
Lever y mm	+14.08
Lever z mm	+5.36
log(R) rad	[−0.0743, +0.3448, +1.5944]
td ms	−1.254
Reprojection px	0.336
cam2
fx px	271.99
fy px	271.94
cx px	323.88
cy px	230.72
k1	0.07207
k2	−0.01495
k3	0.18486
k4	−0.10630
Lever x mm	−75.65
Lever y mm	+14.97
Lever z mm	−30.30
log(R) rad	[+0.9540, −1.3303, −1.0680]
td ms	−1.291
Reprojection px	0.350
cam3
fx px	270.10
fy px	270.63
cx px	319.89
cy px	232.32
k1	0.07858
k2	−0.02786
k3	0.20688
k4	−0.11902
Lever x mm	+80.81
Lever y mm	+16.40
Lever z mm	−29.80
log(R) rad	[−1.3828, +0.9782, −1.3770]
td ms	−1.348
Reprojection px	0.492
IMU
t_ga ms	+2.54 ± 0.20

Limits

• One recording shown, with no independent reference on this page. These values show what MVIS solved, not how close they are to the truth.
• This rig is stitched from three solves, not one joint solve. Cameras seen in two pairs agree to 0.89 mm and 0.34° (cam0) and 0.54 mm and 0.15° (cam1). cam2 and cam3 move 0.79 and 0.52 mm depending on how they are attached.
• The process figure and values for the IMU show the cam0 + cam1 recording only.

How it was run

• Each recording shows the board to one camera pair, so three pairs are solved: cam0 + cam1, cam0 + cam2, cam1 + cam3. Each is camera + IMU with camera intrinsics free and t_ga estimated, on current MVIS, 2026-11-03. cam0, cam1 and all IMU values come from the first pair. cam2 is attached through cam0 and cam3 through cam1.