So, I've tried working with dataset, and it payed off a bit. Now general looks and proportions, work relatively good. While there is still room for improvement, its miles better then first model.
Again, this its still far from perfect tho and there is a huge room for improvement. But still, as said, much better then first model.
Unlike v1 model, this one actually benifits from additional post proccesing.
For preview images, I've used ADetailer on face and hi-rez fix at 0.25 using Remacri.
Besides that really nothing else. You might need to specify poses in your prompt if you want something specific.
Due to the dataset, shoulder boards are messed up as well as armband. I'il try to work with it in the future.
Sometimes there could be "glowing like" effect near the edges. Again thanks to the screencap dataset.
1. 轉载模型僅供學習與交流分享，其版權及最終解释權归原作者。
2. 模型原作者如需認領模型，請通過官方渠道联系海藝AI工作人員進行認證。我們致力於保護每一位創作者的權益。 點擊去認領
