3 ms·
It's trickier than I thought it would be. Here are a few in Degar style I made after training for 2,500 steps. I'd love to hear what you think of them. To my (
by davidbarker 2y ago
It's trickier than I thought it would be.
Here are a few in Degar style I made after training for 2,500 steps. I'd love to hear what you think of them. To my (untrained) eye, they seem a little too defined, perhaps?
https://imgur.com/a/sqsQLPg https://imgur.com/a/sqsQLPg
- vessenes 2y agoYep absolutely nothing like degas well I take that back. I think it picked up some favorite colors/tones. But it has no concept of the materials or poses or composition. So plasticky! Compare to https://images.app.goo.gl/JiDRYNNKUP9tczkQ7 https://images.app.goo.gl/JiDRYNNKUP9tczkQ7
- davidbarker 2y agoI suspect it really needs more training examples. The problem I found when I looked for images to use was that 60% were of dancers, and from past experience, it will end up trying to fit a dancer into every image you create. But of course, there are only a (small) finite number of Degas images that you can train with. A possible solution may be to incorporate artificial images in the training data. So, create an initial LoRA with the original Degas images and generate 500 images. From those generated images, pick the ones that most resemble Degas. Add those to the training set and train again. Repeat until (hopefully) it learns the correct style.
- davidbarker 2y agoOut of curiosity, what do you think of these? https://imgur.com/a/8p7RlMe https://imgur.com/a/8p7RlMe
- vessenes 2y agoSignificantly better - they feel like watercolor more than degas but if that’s flux I am impressed!
- davidbarker 2y agoUnfortunately, not Flux. They're from Midjourney, using a few Degas as a style reference. Whatever they're doing at Midjourney is still impressive. No training needed and a better result.