Quickly extracts a rectangular portion of an image. The tool is configured with an arbitrary coordinate frame and a size in world coordinates. It calculates the axis-aligned bounding box for the area in the input image's pixel coordinates and outputs an image that is a sub-image reference to the input image pixels, if possible. If the specified area is not fully covered by the input image, parts (or all) of the output will be left black.

The difference between cropping and virtual view is that this tool never applies any transformations to the input pixels. The orientation and scale of the output will be the same as that of the input.


The input image.
The size of the rectangle to be cropped in world coordinates. Note that a 0-by-0 size produces a one-pixel image. The actual size of the output image varies depending on frame. If size is undefined (either width or height is NAN), the whole image is passed through without cropping.
A coordinate frame for the rectangle in the world coordinate system.


A cropped output image. This will be a sub-image reference to the input image, if possible.