SP1: Real-Time 3D Stereo Vision Through FPGA-Technology

SP1 Stereo Vision Sensor

The SP1 stereo vision sensor allows you to perform stereo matching on real-time camera imagery, which facilitates depth perception and 3D-reconstruction. Unlike most other approaches to depth sensing, such as structured light or time-of-flight measurements, stereo vision is a purely passive technology which also works in bright daylight.

You can connect two standard USB industrial cameras to the SP1, and thus maintain control over all parameters of the stereo vision setup. Using FPGA-technology and hardware implementations of state-of-the-art algorithms, the SP1 can process the camera images in real-time. The processing result, which is a sub-pixel accurate disparity map (an inverse depth map), is sent out over the gigabit-ethernet port to an attached computer or embedded system. Using our open-source and cross-platform API, this disparity map can be converted into a 3D point cloud.

Example stereo camera

We have put great efforts into optimizing and fine-tuning our algorithms, to ensure that the SP1 delivers high quality depth data at high processing rates. Furthermore, we have made the SP1 easy to use, by integrating a user friendly web interface and by performing automated camera re-calibration. The SP1 is the ideal choice for reliable 3D depth sensing for a diverse set of industrial applications. The FPGA design that drives the SP1 can also be licensed as an IP core.

Depth Measurement Examples

Example left camera image Example depth measurements
Close
Example left camera image
Example left camera image.
Close
Example depth measurements
Depth measurements (disparity map) computed by SP1.
Example left camera image Example depth measurements
Close
Example left camera image
Example left camera image.
Close
Example depth measurements
Depth measurements (disparity map) computed by SP1.
Example left camera image Example depth measurements
Close
Example left camera image
Example left camera image. Taken from the Middlebury 2006 dataset.
Close
Example depth measurements
Depth measurements (disparity map) computed by SP1.

Product Specifications

Hardware Details
Power consumption: < 4 W
Power supply: 5 V DC
Dimensions: 105 × 76 × 36 mm
Weight: 0.25 kg
I/O: USB 2.0 host, Gigabit Ethernet, 2× trigger out, synchronization input (pin shared with trigger 1)
Stereo Matching
Stereo algorithm: Semi-Global Matching (SGM)
Image resolution: Configurable from 320 × 240 pixels up to 1440 × 1440 pixels
Disparity range: Configurable from 32 to 256 pixels
Frame rate: Up to 40 fps depending on image size and disparity range
Sub-pixel resolution: 4 bits (1/16 pixel)
Post-processing: Erroneous disparity removal, noise removal, speckle filtering
Image Rectification
Horizontal displacement: -39 to +39 pixels
Vertical displacement: -39 to +39 pixels
Interpolation: Bilinear

Achievable Frame Rates and Image Resolutions

Image Resolution
640 × 480 800 × 592 1248 × 960 1440 × 1072
Disp.
Range
64 pixels 40 fps 25 fps 9 fps 7 fps
128 pixels 30 fps 20 fps 8 fps 6 fps
256 pixels 15 fps 10 fps 4 fps n/a

Videos

Supported Cameras

The SP1 can be used with our own Karmin 3D stereo camera. In addition a variety of greyscale cameras from FLIR (formerly Point Grey), Basler AG and IDS Imaging Development Systems GmbH are supported. Please follow the link below for a complete list of cameras that are known to be compatible:

Supported Platforms and Standards

ROS HALCON MATLAB MATLAB GenICam GenTL

Models

Model  Description
SP1SP1 stereo vision sensor
Contact Sales

Downloads

Resources

See also