I’m processing a number of photos from exhibitions and concerts. These have dark backgrounds with which I’m perfectly happy as it reduces distractions from the subject. They do need processing in Topaz to denoise and sharpen, but I don’t want to do “Adjust Lighting”, which Autopilot keeps suggesting. Also on the exhibition shots Topaz also insists on trying to do “Recover Faces” on the blank plastic faces of the mannikins.
These are both a real nuisance as Adjust Lighting and Recover Faces are relatively slow models, and it really slows down my workflow to have to wait until autopilot has applied them, delete them, and then wait again to see the effect with the filters I actually want. I’m aware that Topaz is supposed to learn which filters I use for a given photo subject/loo, but there doesn’t seem to be any sign of it learning what I want here.
Unless I’m missing something, there doesn’t seem to be any way to set my preferences so that Autopilot is restricted to certain filters, but others must be selected manually.
Is there any way to do this? If not, can I suggest this as a high-value enhancement?
There are two dimensions to this, which have two different fixes. Either would be good, both would be great.
The first is the ability to set which filters are and are not enabled for autopilot. That functional enhancement would get rid of the problem of “unwanted” filters like Enhance Lighting, or face recovery for mannequins, but would not on its own be a complete solution.
The other is learning how I want to process similar shots. I have just worked through about 40 images from one concert. Each was taken on the same camera, with almost identical exposure, the same dark background, and in most cases similar shapes and lighting on the subject. Any proper “learning” capability would have realised after 2 that I wanted to apply denoise and sharpening (with the Motion Blur model) to each. Topaz resolutely failed to do so.
However, if this is too difficult it should be possible to recognise “similar to the last one he processed” given the above similarities, and just start with the same filters as I manually selected on that last one. This would mimic learning pretty effectively.
I hope this is useful. Please let me know when there’s something to test. If you would like me to do some early testing I have some other similar sets in the queue, please let me know.
Please correct me if I’m misunderstanding, but I don’t understand why I’m unable to disable lighting adjustments for autopilot. I can disable face detection but I’m getting adjusted lighting no matter what I do. Unfortunately for my style of photography, I consider this process to ruin my photos. I realize I can go through and remove it individually, but I edit in batches and this is introducing a slowdown to my workflow that I didn’t have in Photo AI. It seems obvious to me that there should be a preference on each filter to include or disable in autopilot. Hopefully that’s on the roadmap.
We currently do not have the option to turn off Auto-pilot for specific enhancements. However, you can turn off auto-pilot completely in your Preferences and manually apply your preferred enhancement from the enhancement lists.
If you would also like to apply the same filter on more than one image, it will be best to use the batch processing method.
However, your suggestions are duly noted and will be mentioned to the development team as feature requests.
We will update this thread as soon as there is a resolution.
