Emma Watson (JG)版本v0.1 (2010 version) (2010 version)

A commission of @Springbok707.

Emma Watson is a British actress who needs no introduction. For this embedding, the original idea was to capture her 2010 looks, with that famous pixie haircut.

Update: I've now updated the TI to look more like nowadays Emma. Version 3.0 is step 140 of a TI trained on a dataset of 15 images with these settings. It solves mainly some issues with her hair and her eyes present in v2.0.

Curious about my work process? I have summarized it here.

If you want to help me keep creating TIs, please consider buying me a coffee.

Also, I appreciate 5-star ratings if you really like my TIs!

Building a good prompt with my TIs

You're obviously free to experiment, but bear in mind that my TIs are trained with a more or less fixed phrasing, that normally starts with:

"photo of EMBEDDING_NAME, a woman"

So I recommend always starting your prompt like that and then building the rest of the prompt from there. For instance, "photo of (emm4w4ts0n:0.99), a woman as a movie star, modelshoot style, (extremely detailed CG unity 8k wallpaper), photo of the most beautiful artwork in the world, professional majestic oil painting by Ed Blinkey, Atey Ghailan, Studio Ghibli, by Jeremy Mann, Greg Manchess, Antonio Moro, trending on ArtStation, trending on CGSociety, Intricate, High Detail, Sharp focus, dramatic, photorealistic painting art by midjourney and greg rutkowski, (white turtleneck top:1.2), ((movie premiere)), (long skirt), ((standing near a movie theater)), ((paparazzi in the background)), (looking at viewer:1.2), (detailed pupils:1.3), ((closeup portrait:1.1))"

描述:

This isn't the first time I publish a 100-step version... but it certainly isn't usual. I was really surprised during training by how good the image that the AI generated for step 100 looked, so after publishing the TI I decided to test it... And it doesn't look half bad! I'm publishing it more like a curio than anything, cause it's pretty obvious it still needed more training by then. If someone can offer some insights as to why the AI learnt this subject so fast, I'd love to read about it. Just to be clear, the TI was trained under the name "3mwatson2010", so I guess there's no way the AI can relate the vectors of "Emma Watson" already present in the database with the ones that were being trained for this TI. But maybe I'm wrong about that...

训练词语: emw100