3-D vision is the process of reconstructing a 3-D scene from two or more views of the scene. Using the Computer Vision System Toolbox™, you can:
Perform dense 3-D reconstruction using a calibrated stereo pair of cameras.
Reconstruct the scene using an uncalibrated stereo pair of cameras, up to unknown scale.
Compute a sparse 3-D reconstruction from multiple images, using a single-calibrated camera.
|Stereo Camera Calibrator||Estimate geometric parameters of a stereo camera|
|Camera projection matrix|
|Detect checkerboard pattern in image|
|Calibrate a single or stereo camera|
|Generate checkerboard corner locations|
|Correct image for lens distortion|
|Correct point coordinates for lens distortion|
|Object for storing intrinsic camera parameters|
|Object for storing camera parameters|
|Object for storing standard errors of estimated camera parameters|
|Object for storing standard errors of estimated camera intrinsics and distortion coefficients|
|Object for storing standard errors of estimated camera extrinsics|
|Object for storing stereo camera system parameters|
|Object for storing standard errors of estimated stereo parameters|
Specify pixel Indices, spatial coordinates, and 3-D coordinate systems
Estimate the parameters of a lens and image sensor of an image or video camera
Calibrate a stereo camera, which you can then use to recover depth from images.
This example shows how to detect people in video taken with a calibrated stereo camera and determine their distances from the camera.