I would like to have the ability to blend multiple model settings while viewing an image.
Often times I see benefits to 2 models and so I would like to maybe apply 50% of each. Rather than having to process the image both ways and then adjust them as layers in Photoshop, it would be nice to see a blend of 2 models in the preview so that I can play with the adjustments of each model and how much of each to blend before fully processing it. As an example I might want to apply 30% from High Quality and 70% from Text & Shapes because while I might like High Quality I might not want the full effect. Once applied, both models are fully rendered and generated as layers with the desired opacity so that once in Photoshop further adjustment can be applied. Currently, very often I will only use 60% of the high quality mode over the original. Being able to generate 2 models at once (and be able to preview the effect first) would give even more control.
Lastly, it would be nice to be able to preview a single model over the original with varying opacity. So let’s say I choose one model and render it at 50%, it will preview as 50% over the original. Once submitted, it would come in as a layer in Photoshop at 50%, similar to what I described above with multiple models. And with multiple model you could even generate one at 50% and another at 10% which means 40% of the original would show through.
Thanks,
Scott
Hi Scott.
The AI Models were never intended to be used as you describe only as general filters for enhancing images before Upscaling.
Don’t get me wrong, if this method works for you then that’s great continue working that way with Photoshop Layers.
Although, this sounds as though you’re trying to do what is already achievable with Photo AI or at least similar
Photo AI has the ability to stack multiple instances of the same Enhancements and apply different settings to each for different parts of the image for example, Denoise
Or as I like to call it Denoise on a Brush
you can select Denoise and using the Regular Masking Brush apply Standard Denoise to the Highlights with a low Strength for intensity similar to an Opacity if you like then, apply another Denoise Enhancement on top and increase the intensity again think of this as an Opacity for the Midtones and finally a third Denoise selecting Strong and apply a more aggressive denoising for the Shadows I use a similar process for Sharpening or any of the Photo AI Enhancements.
The point being you can do all that, within the Photo AI application
In addition but, not officially supported by Topaz you can also, use Photo AI as a Smart Object/Filter within Photoshop or applying Photo AI Enhancements to individual Layers as a Filter.
Although, Gigapixel can be used for Upscaling and fixing old scans I’m afraid none of the aforementioned Photo AI features above are possible with Gigapixel and never will be available just look at the price difference between the two so, I can’t see Topaz implementing the suggestions you mentioned or Photo AI features into Gigapixel anytime soon
Here’s some information about Photo AI vs Gigapixel
Hope this helps
I’m not sure what you mean that the Gigapixel AI models were never intended to be used this way. If running 2 models and blending them yields a better result than just an individual model then that use is perfectly valid. I am just asking for an easier way to facilitate that. I am speaking specifically of Gigapixel upscaling. Not blending and masking various areas of denoise or sharpening.
If Photo AI has this ability I’ll have to look at it but I have not found PhotoAI to yield the same quality as Gigapixel does for upscaling.
Yes thats true.
I do it via Photoshop.
This is also how the Developement of PhotoAI did start, stacking of filters and a more effective workflow.
Gigapixel has not such a function yet.
Would like to enhance my POI with redefine or recovery and the background via HiFiV2.
