Miss 15, it seems like it was put together in a hurry last weekend, a simple two-in-one layering (layering parameters: 0,0,0,0,0.33,0.66,1,1,1,1,1,1,1,0.1554,0,0,0,0.7489,0.4835,0.2649,0,0,0,0,0,0.1423@0,0,0,0.5,0.5,0,0,0.5, after running dozens of XYZ charts, I believe these are the preferred parameters). The purpose is to maintain the color of Miss 14 as much as possible while using the body of Miss 10 for fusion and repair. As for whether she can demonstrate the generalization ability of Miss 10, I do not know, interested friends please help test it.

        The cover image was lazily copied from previous cues, the process was quite smooth, without turning on facial repair and ad face-editing plugins. The success rate of card drawing should be higher than Miss 13 and 14, similar to Miss 10.

        There are two files, the original color is the initial version generated after layering, with normal skin color, I think it's slightly yellowish, so I adjusted the color slightly, reduced a bit of yellow, and added a bit of contrast. The output images are the same, the only difference is in color, I uploaded the original color maybe it would be useful for me in some brushing scenes, usually you can just download the color adjustment.

      Miss 14's facial collapse happened because it's easy to produce full-body images, with the body occupying a relatively small percentage in the picture, while Miss 15's body proportion is slightly larger, so sometimes when you want a full-body image, you may need to add the fullbody cue words.