Functional Specifications#

This topic details the functional specifications of the Stereo mini camera series.

Vendor Identifier (VID) and Product Identifier (PID)#

Platform and System Requirements#

Cameras of the Stereo mini series connect to the host computer using USB or Ethernet, which is compatible with various platforms and system requirements.

USB#

GigE#

STM-951g#

STM-955g#

Camera System Framework#

USB#

GigE#

Image Data Stream#

USB#

USB cameras of the Stereo mini series provide high-quality, multi-resolution depth stream data as well as high-definition color stream data. The camera outputs depth stream data in Y16 format. The color stream data output by the camera is in MJPEG/YUYV format. The SDK supports output in MJPEG/YUYV/RGB8/BGR8/RGBA8/BGRA8/Y16 formats. The camera outputs IR image data in Y8 format, and the SDK also supports outputting data in Y16/Y12 formats at maximum resolution.

Output Data Streams (USB3.0): STM-501u and STM-502u#

Output Data Streams (USB2.0): STM-501u and STM-502u#

Output Data Streams (USB3.0): STM-951u and STM-952u#

Output Data Streams (USB2.0): STM-951u and STM-952u#

GigE#

GigE cameras of the Stereo mini series provide high-quality, multi-resolution depth stream data, as well as high-definition color stream data.

The STM-951g outputs depth stream data in Y16 format. The color stream data output by the camera is in MJPEG/YUYV format. The SDK supports output in MJPEG/YUYV/RGB8/BGR8/RGBA8/BGRA8/Y16 formats and also supports output in RAW16 format at maximum resolution. The camera outputs IR image data in Y8 format, and the SDK also supports outputting data in Y12 format.

The STM-955g outputs depth data in Y16 or RVL format, color images in I420 or MJPEG format, and IR images in Y8 format. The software SDK also supports RGB/BGR/BGRA formats for color streams.

Output Data Streams (Gigabit Ethernet): STM-951g#

Output Data Streams (Fast Ethernet): STM-951g#

Output Data Streams (Gigabit Ethernet): STM-955g#

Field of View#

Definition of Depth Field of View#

The image below shows the depth camera's field-of-view, or the angles that the sensors "see". The IR cameras are used for illustration.

Depth field of view (depth FOV) at any depth (Z) can be calculated using the following equation:

其中：

cx = X-direction image coordinates of the principle point of the depth image

fx = Depth camera focal length

cy = Y-direction image coordinates of the principle point of the depth image

fy = Depth camera focal length

width = Depth image width

height = Depth image height

Depth active H-FOV = Left IR H-FOV

Typical Depth Intrinsics#

The following tables list typical depth intrinsics for Stereo mini cameras.

STM-501u and STM-502u#

Baseline: 50 mm

STM-951u and STM-952u#

Baseline: 95 mm

STM-951g#

Baseline: 95 mm

STM-955g#

Baseline: 95 mm

Overview of Stream FOV#

USB#

GigE#

FOV Illustrations#

STM-501u / STM-502u#

Aspect Ratio	Depth FOV Before D2C	Depth FOV After D2C
16:10		不适用
16:9
4:3

STM-951u / STM-952u#

Aspect Ratio	Depth FOV Before D2C	Depth FOV After D2C
16:10
16:9
4:3

STM-951g#

Aspect Ratio	Depth FOV Before D2C	Depth FOV After D2C
16:10
16:9	不适用
4:3

STM-955g#

Aspect Ratio	Depth FOV Before D2C	Depth FOV After D2C
16:10
16:9	不适用
4:3

Depth to Color Alignment#

Depth to color (D2C), a pixel-by-pixel geometric transformation of a depth image, results in the spatial alignment of a depth image with its corresponding color image through the D2C transformation. This makes it possible to locate any pixel of a color image based on its image coordinates in the depth image after D2C by the same image coordinates.

The depth information of the color pixel can be located in the depth image after D2C by using the same image coordinates. A depth image of the same size as the target color image after D2C is generated, and the image content is the depth data in the color camera coordinate system. In other words, a depth image that has been "taken" using the origin and size of the color camera is reconstructed, where each pixel matches the coordinates of the corresponding pixel of the color camera.

D2C Alignment by Software: STM-501u / STM-502u#

D2C Alignment by Hardware: STM-501u / STM-502u#

D2C Alignment by Software: STM-951u / STM-952u#

D2C Alignment by Hardware: STM-951u / STM-952u#

D2C Alignment by Software: STM-951g#

D2C Alignment by Hardware: STM-951g#

D2C Alignment by Software: STM-955g#

D2C Alignment by Hardware: STM-955g#

Minimum Z Depth#

The minimum Z depth is the minimum working distance of a depth camera, i.e., the closest detectable planar distance from the camera's depth origin. This distance is determined by the disparity search range and image resolution.

The STM-955g adjusts its minimum working distance by configuring depth offset values. The achievable minimum distance may also vary across different Preset modes. The camera provides flexible configuration options to adapt to diverse application needs.

STM-501u / STM-502u#

STM-951u / STM-952u#

STM-951g#

STM-955g#

Coordinate System#

For cameras of the Stereo mini series, the plane where the 1/4 screw hole is located is defined as the bottom side, the glass cover surface is the front side, and the RGB module is positioned to the left of the LDM module.

The origin of the IMU coordinate system is situated at the physical sensor center point. The accelerometer and gyroscope coordinate systems are located at the back of the left IR. The positive X-axis of the coordinate system points to the right, the positive Y-axis points downwards, and the positive Z-axis points forwards.

The origin of the depth image coordinate system is at the optical center of the left IR module, while the origin of the color image coordinate system is at the optical center of the RGB module. The direction of the coordinate systems is the same: the positive X-axis points to the right, the positive Y-axis points downward, and the positive Z-axis points forward. The depth camera coordinate system origin is the default origin of the 3D camera, with coordinates (0,0,0). The reference positions of the depth origin, color origin, and IMU origin in the 3D camera coordinate system are shown in the Coordinate System Schematic table below.

Coordinate System Position Reference#

Coordinate System Schematic#

	STM-501u / STM-502u	STM-951u / STM-952u / STM-951g	STM-955g
IMU
Depth
RGB

Coordinate System Position Reference#

The relative coordinate system relationship between the installation positions and Depth/Color/IMU.

STM-501u / STM-502u#

STM-951u / STM-952u#

STM-951g#

STM-955g#

Camera Start Point Reference#

The camera start point, or ground zero datum can be described as a start point or plane with depth = 0. The distances of the depth/RGB/LRM zero point relative to the front cover glass of the camera are listed in the table below.

Streaming Mode#

The Stereo mini series offers users flexible methods for acquiring IR, Depth, and RGB image data, with the streaming mode being the most common approach. In this mode, users set a target frame rate, resolution, and image format for each data type before sequentially activating the corresponding data streams. The camera then captures and outputs image data at the user-defined target frame rate, resolution, and image format.

You can select a specific frame rate for the current scene from predefined options (5 fps / 6 fps, 10 fps, 15 fps, 30 fps, 60 fps, 90 fps, and 100 fps) based on the camera's currently configured depth mode and resolution and capture image data at that frame rate.

Triggering Mode#

The Stereo mini series supports image data acquisition methods based on specific frame rates as well as a free-running triggering mode that supports arbitrary frequencies. In this mode, the camera waits for an external input trigger signal and only completes an image data acquisition after receiving a valid external trigger signal, as configured by the camera. The camera then continues to wait for the next external trigger signal. As there is no set time limit between two consecutive triggers, but only a single acquisition time greater than that of the camera, it is possible to control the time interval between two consecutive triggers to achieve any desired frequency. This allows for passive acquisition of image data. Depending on your camera's interface, the camera can be triggered in the following ways:

• USB camera models: Software signal sent by the host computer via USB command or an external device through the 8-pin synchronous interface. This facilitates a passive triggering mode at any frequency.
• STM-951g: Software triggers via network commands (host-initiated) and hardware triggers through M8 A-coded sync interface
• STM-955g: Software triggers via network commands (host-initiated) and hardware triggers through M12 A-coded sync interface

In free triggering mode, the camera's IR, Depth, and RGB fixed frame rates must be set to a uniform value of 5 fps / 6 fps, 10 fps, 15 fps, 30 fps, 60 fps, or 90 fps (depending on the camera model) as required. This is necessary to determine the minimum time interval between two consecutive active triggers. The following tables show the relationship between the fixed frame rate, the minimum time interval, and the upper frequency limit for passive triggering.

In summary, the camera will only respond to trigger signals within the allowed range. This means that the trigger frequency can be any value within the valid frequency range for passive triggering.

USB Camera Models and STM-951g

STM-955g

Multi-Camera Synchronization#

For a multi-camera use case, one camera can be initialized as primary, and the rest configured as secondary. Alternatively, an external signal generator can also be used as the primary trigger with all cameras set to secondary mode. When applying an external sync pulse, the HW SYNC input requires a 100-microsecond positive pulse at the nominal camera frame rate, e.g., 33.33 ms for a 30 Hz frame rate. Inputs are high impedance, 1.8 V for USB and 3.3 V for GigE models CMOS voltage levels. However, it's important to make sure to use a high-resolution signal generator. The frequency of the signal generator needs to exactly match the sensor frame rate. For example, if the sensor is set up as 30 fps, the real frame rate may be 30.015 fps. You may need to use an oscilloscope to measure the real frame rate and configure the signal generator to the same frequency. For this reason, it may be better to just use one additional camera as the primary sync signal generator.

The following figure shows a mechanical diagram of a multi-camera setup.

Advantages of multi-camera setup：

• Increase camera coverage in a given space and fill in the occlusions where a single camera may have blind spots
• Capture multiple images of the same scene and scan objects from different angles
• Increase the effective frame rate to above 30 fps

Using the 8-pin connector and matching cable, a multi-camera and multi-sensor network can be designed. Follow the instructions in the SDK.

Multi-camera frame synchronization in two topologies is supported, including depth image synchronization and RGB image synchronization (time difference ≤ 5 ms, when auto exposure off), using the multi-camera synchronization function.

• Star Topology
  
• Daisy-Chain Topology
  

Synchronization Interfaces#

USB#

The synchronization point placements are shown in the following drawings.

GigE#

The synchronization point placements are shown in the following drawings:

Pin numbering of the power connector (identical for both camera models):

Camera Functions#

Depth Camera Functions#

The Stereo mini series exposes the following depth image settings.

Color Camera Functions#

The Stereo mini series exposes the following Color image settings.