This document details the specifications and some design details of the ORBBEC® Femto Mega 3D camera product for developers’ understanding and use.
Terminology
The following abbreviation terms are used throughout.
- NFOV (Narrow Field of View depth mode)
- WFOV (Wide Field of View depth mode)
- FOV (Field of View)
- FPS (Frames per Second)
- IMU (Inertial Measurement Unit)
- FoI (Field of Interest)
Product Dimensions and Weight
The Femto Mega device dimensions and weight are:
- Dimensions: 115.30 × 143.5 × 40.3mm
- Weight: 560g
For more information, please refer to the github https://github.com/orbbec/OrbbecHardware
Operating Environment
Femto Mega is intended for use by developers and commercial deployments operating under these environmental conditions:
- Temperature: 10-25°C
- Humidity: 8-90% (non-condensing) relative humidity
Note: Operation outside these conditions may result in device failure and/or abnormal operation. These environment conditions pertain to the ambient environment immediately surrounding the device while in operation under all use cases. If used with an external enclosure, effective temperature control and/or other heat dissipation solutions are recommended to maintain operating conditions within these ranges. A thermal channel is designed between the front face and rear cannister of the device. Do not obstruct this thermal channel when using the device.
Depth Camera Supported Operating Modes
Femto Mega integrates the 1 megapixel Time-of-Flight (ToF) depth camera designed by Microsoft. The depth camera supports the following modes as described:
Mode | Resolution | FoI | FPS | Operating range* | Exposure time |
NFOV unbinned | 640×576 | 75°x65° | 5、15、25、30 | 0.5 – 3.86 m | 12.8 ms |
NFOV 2×2 binned (SW) | 320×288 | 75°x65° | 5、15、25、30 | 0.5 – 5.46 m | 12.8 ms |
WFOV 2×2 binned | 512×512 | 120°x120° | 5、15、25、30 | 0.25 – 2.88 m | 12.8 ms |
WFOV unbinned | 1024×1024 | 120°x120° | 5、15、25 | 0.25 – 2.21 m | 20.3 ms |
Passive IR | 1024×1024 | N/A | 5、15、25、30 | N/A | 1.6 ms |
*850nm at 15% to 95% reflectivity, 2.2 μW/cm2/nm, standard deviation of random error ≤ 17 mm, typical system error < 11 mm + 0.1% of distance (no multi-path interference). Depth can be provided beyond the indicated operating range depending on object reflectivity.
Color Camera Supported Operating Modes
The native operating modes are listed below:
RGB Camera Resolution (HxV) | Aspect Ratio | Format Options | Frame Rates (FPS) | Nominal FOV (HxV)(post-processed) |
3840×2160 | 16:9 | MJPEG/YUY2/H.264/H.265 | 5、15、25 | 80°x51° |
2560×1440 | 16:9 | MJPEG/YUY2/H.264/H.265 | 5、15、25、30 | 80°x51° |
1920×1080 | 16:9 | MJPEG/YUY2/H.264/H.265 | 5、15、25、30 | 80°x51° |
1280×720 | 16:9 | MJPEG/YUY2/H.264/H.265 | 5、15、25、30 | 80°x51° |
1280×960 | 4:3 | MJPEG/YUY2/H.264/H.265 | 5、15、25、30 | 65°x51° |
Note: The SDK can provide color images in BGRA pixel format. This is not a natively supported mode by the device and will result in increased CPU load. The host CPU is used to convert from the MJPEG images received from the device.
Depth Sensor Raw Timing
Depth Mode | IR
Pulses |
Pulse
Width |
Idle
Periods |
Idle Time | Exposure
Time |
NFOV Unbinned
NFOV 2×2 Binned WFOV 2×2 Binned |
9 | 125 us | 8 | 1450 us | 12.8 ms |
WFOV Unbinned | 9 | 125 us | 8 | 2390 us | 20.3 ms |
Camera Fields of View
The image below shows the depth and RGB camera fields of view (what the sensors see). The 16:9 mode RGB camera is shown below.
This depicts the camera fields of view at a distance of 2000 mm straight ahead.
Note: For depth in NFOV mode, RGB camera pixel overlap is better in 4:3 mode than 16:9 mode.
Inertial Measurement Unit (IMU)
The embedded Inertial Measurement Unit (IMU) is a LSM6DSMUS, containing accelerometer and gyroscope. 6-axis accelerometer & 6-axis gyroscope, data format is float, frequency range is 50-2000Hz.
Indicator Lights
An indicator light on the front of the device signals the working status of the device, on by default and can be manually switched off.
A status LED on the back of the device indicates:
When the light is | It means |
Solid white | Device is on and working properly. |
Flashing white | Device is on but doesn’t have a USB 3.0 data connection. |
Flashing amber | Device doesn’t have enough power to operate. |
Powering the Device
The device can be powered in two ways:
- Using the supplied power adapter. The power connector has an outer diameter of 5.5mm and an inner diameter of 2.1mm.
- Power and data transfer via a Type-C to Type-C cable.
- Power over Ethernet (PoE) supply + Gigabit Ethernet data transmission
Note: The supplied power cable is a USB Type-A to barrel jack. Use this cable along with the supplied wall adapter. Two standard USB Type-A ports do not provide sufficient power for this device.
The USB cable is very important, a quality cable is recommended and functionality should be validated before remote deployment of the unit.
Tips for a good Type-C to Type-C cable:
- USB certified cable – must support power and data transfer
- Passive cables should be less than 1.5m length, use active for longer
- Cable needs to support at least 1.5A current, otherwise external power required
- Test cable:
- Connect device to host PC with cable
- Verify all devices appear correctly in Windows Device Manager. Depth and RGB cameras should appear as shown in example below.
- Verify cable reliably streams all sensor data in K4A Viewer with these settings:
- Depth camera: NFOV unbinned
- RGB camera: 2160p
Power Consumption
Femto Mega peak power consumption is 16W with DC power + Type-C data transfer; 14W with Type-C power + Type-C data transfer.17W with Power over Ethernet (PoE) .Power consumption is use case dependent.
Calibration
Femto Mega devices are calibrated at the factory. Calibration parameters for the visual sensors and inertial sensors can be programmatically queried via the Sensor SDK.