GetVideo
GetVideo(srcClip, timeOffset, destImage)
Additional Notes
Notes on offset values
The value specified by timeOffset refers to srcClip's current inpoint. Strangely, to get this frame, GetVideo needs to call the clipoffset as a negative number. The actual first frame of the clip is twice the negative value of the offset. This is why the GetVideo scripts usually subtract the clip's offset value from the frame constant to call a copy of the currently visible video frame.
Current Frame = -clipoffset + frame/fieldprocessing
Offset values in "image wells"
The clip offset works differently when you're accessing a frame of video from video that has been dropped into an "image well". In this case, you must add the offset rather than subtract to get the correct video frame.
This one is confusing.
GetVideo will scale the requested frame to fit whatever size the destImage target is. This can be useful, but also infuriating. If GetVideo is placed into an image buffer equal in size to an interlaced source clip, the fields of the video will be squished, which is a slightly muddy looking result.