< Previous | Contents | Manuals Home | Boris FX | Next >
Selecting the Detail Mode
To track, SynthEyes matches up the image on a particular frame to the reference frame. Most of the time, features being tracked within the planar tracking mini-view occupy a fairly large portion of it, and it is comparatively easy to get the images to align.
Suppose we're tracking a red dot on a green screen. If we make the planar tracker only somewhat larger than the red dot, when we compare the reference and new images, the red dots will overlap a bit, and it will be easy to get them to match up exactly.
Suppose, however, that we make the planar tracker much larger than the red dot. When we compare the two images, odds are that the dots won't overlap at all. If we shift the images around a little, there is little change: there's a red dot compared to a green background, and a green background compared to a red dot. That's still the story if you shift the images around a bit. It takes some effort to figure out how to get them to align up.
That's what the Detail Mode dropdown is for. Normally, you can run in Low detail mode, for large features.
If there is only fine detail in the tracked image, as in our small red dot example, you should switch to High Detail mode. The High Detail mode requires more computation, but is essential to handling fine patterns, such as thin text or a large area with a few small marks.
The Medium detail is a combination of the two, rather than an intermediate level of processing. In Medium mode, the Low detail processing is performed, then the High detail processing to further refine it. This is worthwhile with a sloppy large pattern and some more precise fine detail that might be problematic in its own right.
So the Medium detail mode can take the most computation of all!
©2024 Boris FX, Inc. — UNOFFICIAL — Converted from original PDF.