Miu Iruma ( Danganronpa ) [SDXL]版本V1 (ID: 879944)

Miu Iruma from the Danganronpa V3. I trained on both her normal V3 outfit and her bikini from Danganronpa S.

IMPORTANT: This sdxl lora is trained on the same data as my original flux Miu Iruma lora. In short, this sdxl lora suffers a bit from being designed for a different system. Though not terribly. The most significant issue is that it doesn't take style directions very well. Specifying danganronpa 3, danganronpa the animation, and danganronpa sprite style, etc will have an effect. Just less of an effect than the flux lora. The sprite style in particular has some big issues. When you use sprite style it'll usually just make things a bit flatter. In retrospect, I didn't train on enough images where the full body sprite of Miu was actually seen in the background. If I ever made an update I'll rectify that.

Unique features: The main point I try to stress in my Danganronpa loras is authenticity to the source material. It's fairly easy to get the basics of a character's style or clothing down. Characterization is a lot harder. Models have their own concepts of what a scared, or happy, or sad, or angry expression or stance would be. But a character in something like Dananronpa has a lot of careful attention put into ensuring their designs make all of those states unique. I go through every sprite, every CG, every official image I can find to caption and include them in the dataset in hope of best representing that. Whether it works out is another question, but it's what I'm aiming for and why these might be a little odd.

Prompt style: I trained using natural language for the captions rather than tags so natural language. So natural language prompts will give you the best results.

Models: So far I'm seeing the best results with Neta Art XL and AAM XL (Anime Mix).

Specific prompts: I trained on a few specific image types which you can replicate, to varying levels of success. I tried to include a range of the prompt options in the images so you can easily compare and contrast. But these are the major options.

Danganronpa the Animation: Use 'This is an anime screencap from Danganronpa The Animation' in the prompt to emulate the art style from the anime series Danganronpa the animation.

Danganronpa 3 (anime): Use "This is an anime screencap from Danganronpa 3."

Sprites from Danganronpa: Mention 'danganronpa sprite style' in the prompt to try emulating the art style of Danganronpa. For example, 'this image is in the danganronpa sprite style'. Despite the name, I trained on full screenshots, with and without the GUI, with this description along with character sprite rips. So in theory it should be able to 'danganronpafy' a full image.

Danganronpa S GUI: You can give it a try with this. "The image uses danganronpa S GUI style with the danganronpa sprite style. A speech bubble is in the bottom of the image and it says "This is test text!" The speech bubble is in front of Miu Iruma. There's a box with a blue background with white stars, and a grey arrow pointing right, beside the speech bubble. The text in the box says "Miu Iruma" and it is next to the speech bubble."

Danganronpa V3 GUI: This will be less reliable than the Danganronpa 3 GUI: To the left of the image there is a yellow bar containing the text 'Music Track Name'. The digital text in yellow on the bar is in danganronpa V3 GUI style. There is a black bar with white text reading 'Chapter 3 Level 72'. In the bottom left corner of the image, there is another yellow bar with an angled black line down the left side, which reads 'Miu Iruma' using a danganronpa V3 GUI style font. Below that, there is a white speech bubble using the danganronpa V3 GUI style containing speech that reads "Miu's Text" In the corner of the image, there is another danganronpa V3 GUI style text button saying 'AUTO.' The image is in Danganronpa sprite style with Danganronpa V3 GUI elements.

Clothing: You can try different clothing options with prompts like the following:

Standard outfit: Miu Iruma wearing her uniform from Danganronpa V3. Miu Iruma has waist-length strawberry blonde hair, ice-blue eyes, brown goggles, and dark pink seifuku uniform. She has a white sailor collar and sleeve cuffs striped with a pale blue line and pale blue bow at her collar. Her cleavage is exposed. A crest's on the shoulders of her sleeves. She wears a short, dark pink pleated skirt, vertically striped off-white knee-high socks, black heeled boots. Her accessories include a black choker, two rubber collars that resemble barbed wire, black studded fingerless gloves, a pin of the mouse mascot on her collar, black straps connected with gold studs wrapping her left upper arm, and black straps around her torso and legs. Her long black boots nearly reach her knees.

Swimsuit: Miu Iruma has waist-length strawberry blonde hair that flows in waves around her back, with two short locks of hair sticking up at the top of her head, sweeping to her right. Her bangs are long and wispy, swept somewhat to her left. She has ice-blue eyes. Her accessories include a black choker and two rubber collars that resemble barbed wire around her neck. Miu Iruma wearing a leopard-print bikini. She is wearing a black choker necklace.

Locations: I didn't try to train on any specific locations, but some of them appeared enough in the training data that you might be able to improve your results by specifying areas from Hope's Peak or Jabberwock island. Even just mentioning them might be enough to help push things in the right direction.

Training material: Every official Miu Iruma sprite and image from Danganronpa V3 and Danganronpa S. Additionally I found any additional official images and higher quality fan art. I also used a large amount of the backgrounds from Danganronpa 1,2, S and to a lesser extent V3. Finally, I used images from Danganronpa the Animation and Danganronpa 3 to allow for the style choices. However, since Miu isn't in those shows it's less exact than it might be for other characters like Makoto.

描述:

训练词语: