NAME

blind-crop - Extract subframes for all frames

SYNOPSIS

blind-crop [-s | -S | -t] width height left top

DESCRIPTION

blind-crop reads a video from stdin and prints a cropped version of the video to stdout. The new video with have the specified width and height, and will have the same length as the input video. The left-most pixels of the subvideo will be left pixels rightward of the left-most pixels of the input video. The top-most pixels of the subvideo will be top pixels downward of the top-most pixels of the input video.

The selected subvideo may not extend beyond the input video.

OPTIONS

-s: Instead of resizing the video, set alpha (and all colour parameters) to 0 on each pixel outside the selected region.
-S: Instead of resizing the video, set alpha (and all colour parameters) to 0 on each pixel inside the selected region.
-t: Instead of changing the width and height of the output video, put the subvideo side-by-side. The subvideo's left-most pixel will be positioned left pixels rightward of the output video's left-most pixel, and the subvideo's top-most pixel will be positioned top pixels downward of the output video's left-most pixel.

REQUIREMENTS

blind-crop requires enough free memory to load two full frames into memory, one of the size of the source video's frames, and one of the size of the target video's frames. However, if -s or -S is used, only memory for one full frame, of the size of the source video's frames, are required. A frame requires 32 bytes per pixel it contains. blind-crop has not been optimised for memory usage, but instead for code simplicity.

AUTHORS