Text to Actor workflow?

AI Studio General
1

I’m exploring switching to AI Studio. My current workflow is:

  1. Text prompts using FLUX and Nano Banana until I get a good base image.
  2. Using that base image, I generate a dozen expressions. I keep generating until I get a dozen good ones. I mix FLUX and Nano Banana to get a good mix.
  3. I use the dozen+ best images to train a LoRA
  4. I then generate a bunch of test images to validate the LoRA. I go backwards and refine as needed, rinse and repeat until LoRA is correct.
  5. Once validated it becomes an actor which I use in generating scenes.

In AI Studio, it seems like I need need all the images shown in the Create Actor panel, but it feels like I’m missing step 2 above? EG It is difficult with the Image Gen prompt to create the set of image slots shown. Am I doing this wrong? What is the correct workflow?

2

The images are generated by using an actual 3D character. There is an iClone template project with a dummy character that has the required positions and expressions for the character definition. You replace the dummy with the desired character and render out into 14 images which are then loaded into the slots in AI Studio. The idea is that that image set is suffient and it has the advantage that you don’t have to go through the LoRA training process.

I don’t have iClone in front of me but that’s the idea.

3

Ah ok I see. I suppose I can use my existing LoRAs to generate the required images and then switch to AI Studio, will give that a try. Cheers