Model
Trends
.ai
Model
Trends
.ai
les meilleurs modèles d'IA open source
Exécuter en local
See-through_body_v1.safetensors
218 MB · SafeTensor
Les téléchargements sont servis par Civitai. Certains créateurs demandent d'abord une connexion à Civitai.
AD
<see-through-body-xl>
À propos de ce modèle
STABLE DIFFUSION XL 1.0 · LoRA
Créé par
PressWagon
sur Civitai
by PressWagon - See-through body | Experimental v.1 normal
Notes du créateur
Anima v2.4 The v2.4 dataset trained for Anima. Results are ok. Works better with a hires/upscale step. v2.4 Thanks to for the tip, it funded a great deal of experimentation. This version finally produces results I like better than v1 but I am curious to hear from you all how it compares in your use. Prompting specific features at full strength, such as facial expressions, can overpower the see-through effect. Reducing prompt weight, e.g. "(smile:0.7)", can produce results that successfully interpolate between both concepts. v2.3 I have tried without success to improve the model. This version lacks some of the older artifacting but is less consistent producing the effect. If I am going to spend all this buzz training bad models I might as well release one of them for you to try. Maybe you will find a use for it. v1 - Large The same dataset as v1 - Normal on a lora that is twice as large. Results seem promising, but your feedback would be appreciated. v1 - Normal (Recommended) With an enhanced dataset the results are improved. There is still a strong style influence which biases the results towards an anime style and backgrounds exhibit overfitting. Effect works best when the background is a repeating pattern (like stone walls, concrete paths) or includes elements that extend across the subject (windows, edges etc). Although it was not tagged for this at all, adding "see-through head" may improve the transparency of the head when used in addition to "see-through body". Adding tags like "see-through clothes" to the negative can get a more invisible girl style. Some issues may also be solved by adding "tattoo" to the negative. v0 Proof of concept: This is a proof of concept for training a concept on synthetic and modified imagery. The initial dataset was very small and the resulting model is overbaked but shows signs of concept comprehension. Method Explainer v0 and v1 were trained entirely on synthetic data created by generating a subject and a background independently and then compositing a semi-transparent subject over a background using image editing software. This approach has an obvious downside, that the two components are not well integrated. How then to create a full body subject oriented correctly to the viewing angle of the background? My solution was to turn to my favourite tool, the lineart controlnet. The initial prompt was: 1girl, standing, full body, turning around, looking back, long blonde hair, red jacket, blue jeans, park, grass, bench, concrete path, masterpiece, best quality, amazing detail, high definition Which could then be modified for the isolated subject: 1girl, standing, full body, turning around, looking back, long blonde hair, red jacket, blue jeans, simple background, masterpiece, best quality, amazing detail, high definition and a background prompt: no humans, park, grass, bench, concrete path, masterpiece, best quality, amazing detail, high definition Masking out the subject becomes much easier with a simple background. Then it can be superimposed as a transparent subject over a fully detailed background. Even using a model that didn't understand "see-through body" we were still able to create synthetic training data so that it could learn it. Pretty neat! Before this I had no idea if this was a viable approach. The results, while not incredible, are better than I was expecting.
Mots déclencheurs
see-through body
784 téléchargements · 141 j'aime sur Civitai
SHA256
58631D3353C36A5919FDB1D901536AA75AA6BDA3371C268A01BC2C5A69F95F62
ID de modèle ModelTrends.ai
#15762
Signaler le contenu
Rendus d'exemple
Prompts et réglages partagés par les personnes qui ont réalisé ces rendus sur Civitai. Choisissez un rendu pour voir comment il a été fait.
Réglages
Sampler
dpmpp_2m_sde_karras
Étapes
30
Guidance
5
Seed
857643285431197
Taille
896x1152
Prompt
1girl, see-through body, school girl, rooftop, looking up, dissolving, smoke aura, one arm raised, glowing, sad, masterpiece, best quality, amazing detail, high definition
Prompt négatif
bad quality, worst quality, worst detail, text, patreon, signature, steam, young,
Réglages
Sampler
DPM++ 2M
Étapes
30
Guidance
5
Seed
2158110827
Taille
832x1216
Clip skip
2
Prompt
traditional media, watercolor, 1girl, solo, honma meiko, soft smile, looking back, waving, see-through body, see-through head, train, motion blur, outdoors, railroad crossing, train tracks, dramatic shadows, soft gradients, abstract, abstract composition, cinematic contrast, beautiful, melancholic, masterpiece, best quality, amazing quality, very aesthetic, high resolution, ultra-detailed, absurdres <lora:t86artstyle_water_collage:1.0> <lora:Honma_Meiko_from_Anohana_IL:1> <lora:train_in_background_il_1_d16:1.0> <lora:See-through_body_v1:1.0> <lora:lighting_temp_slider1.0:2.5>
Prompt négatif
nsfw, uncensored, see-through clothes, tattoo, worst quality, low quality, normal quality, lowres, jpeg artifacts, artistic error, multiple views, signature, watermark, artist name, bad hands, extra digits, extra legs, displeasing, worst aesthetic, old, early, simple background, simple shading, simple drawing, simple coloring, DeepNegative_xl_v1, zPDXLxxx-neg
Réglages
Sampler
DPM++ 2M
Étapes
25
Guidance
3.5
Seed
1611339819
Taille
1280x1856
Clip skip
2
Poids utilisé
0.9
Prompt
masterpiece, best quality, amazing quality, very aesthetic, absurdres, newest, no humans, duck, cute animal, solo, detailed eyes, red eyes, wings up, open beak, zombie, towards to viewer, dark room, indoors, hospital, dutch angle, blood on floor, particles, ((fog:1.2)), blurry background, broken window, broken glass, dynamic light, see-through body, wall-eyed cute, chibi, full body,
Prompt négatif
lowres, bad anatomy, bad hands, error, cropped, extra digits, missing finger, fewer digits, film grain, blurry, jpeg artifacts, adversarial noise, scan artifacts, scanlines, worst quality, low quality, low score, bad score, average score,
Réglages
Sampler
DPM++ 2M SDE Karras
Étapes
28
Taille
2432x1664
Clip skip
2
Prompt
1girl, cybernetic sniper, looking ahead, see-through body, cybernetic bodysuit, hooded cloak, hood up, high collar, goggles, gas mask, cybernetic limbs, futuristic implants, (bio-mechanical fusion:1.2), neon circuit details, aiming, on one knee, holding sniper rifle, thigh belt, porch, gloves, perched on rooftop ledge, rooftop edge sniper position, rubble, sky, (smoky:1.3), (disintegration effect:0.8), hazy air, gradient, (watercolor \(medium\):1.2), asymmetrical clothes, asymmetrical legwear, asymmetry, outdoors, detailed scenery, intricate background, highly detailed background, cinematic background, soft lighting, vibrant colors, (glitch:0.8), cyberpunk, science fiction, future, best quality, amazing quality, very aesthetic, absurdres, newest, cinematic atmosphere, <lora:See-through_body_v1:0.9>
Prompt négatif
lowres, (worst quality, bad quality:1.2), bad anatomy, bad hands, (missing fingers, extra fingers:1.2), fused fingers, long fingers, mutated hands, deformed, disfigured, poorly drawn hands, sketch, jpeg artifacts, text, signature, watermark, old, oldest, 3d, realistic, sweat, tears, steam, fire, bipod
Réglages
Sampler
DPM++ 2M SDE Karras
Étapes
28
Taille
2432x1664
Clip skip
2
Prompt
1girl, out of frame, dynamic angle, from side, see-through body, bodysuit, hooded cloak, hood up, high collar covering mouth, mechanical eye, cybernetic limbs, futuristic implants, bio-mechanical fusion, neon circuit details, aiming, on one knee, holding sniper rifle, thigh belt, porch, gloves, on rooftop, rubble, (smoky:1.3), (disintegration effect:0.8), hazy air, gradient, (watercolor \(medium\):1.2), asymmetrical clothes, asymmetrical legwear, asymmetry, outdoors, detailed scenery, intricate background, highly detailed background, cinematic background, soft lighting, vibrant colors, (glitch:0.8), cyberpunk, science fiction, future, best quality, amazing quality, very aesthetic, absurdres, newest, cinematic atmosphere, <lora:See-through_body_v1:0.9>
Prompt négatif
lowres, (worst quality, bad quality:1.2), bad anatomy, bad hands, (missing fingers, extra fingers:1.2), fused fingers, long fingers, mutated hands, deformed, disfigured, poorly drawn hands, sketch, jpeg artifacts, text, signature, watermark, old, oldest, 3d, realistic, sweat, tears, steam, pov hands,
Réglages
Sampler
dpmpp_2m_sde_karras
Étapes
30
Guidance
5
Seed
857643285431197
Taille
896x1152
Prompt
1girl, solo, serious, elf, colored skin, leather armor, dark fantasy art, sneaking, see-through body, crouching, holding knife, ornate dagger, night, knee boots, stone wall, tree, fantasy town, masterpiece, best quality, amazing detail, high definition,
Prompt négatif
bad quality, worst quality, worst detail, text, patreon, signature, young,
Réglages
Sampler
dpmpp_2m_sde_karras
Étapes
30
Guidance
5
Seed
857643285431198
Taille
896x1152
Prompt
1girl, solo, adventurer, leather armor, brown tunic, examining body, scared, ?!, (slimification:1.2), blue slime, ponytail, discarded backpack, slime legs, slime arms, see-through body, stone walls, against wall, dungeon, masterpiece, best quality, amazing detail, high definition,
Prompt négatif
bad quality, worst quality, worst detail, text, patreon, signature, steam, young, see-through clothes,
Modèles d'IA pour générer image, vidéo et texte
see through body xl est un LoRA pour la famille Stable Diffusion XL 1.0, référencé sur ModelTrends.ai, un catalogue en lecture seule de modèles d'IA open source pour l'image, la vidéo et le texte. Comparez-le aux autres modèles Stable Diffusion XL 1.0, consultez son score de chaleur pour suivre sa tendance et ouvrez-le sur PirateDiffusion ou BitVector pour l'essayer.
Parcourir par famille de modèles
Voir tous les modèles