Virtual Try-on
BETA. Backend V1 feature. Price changes still pending.
![Fine-tuning training images](/assets/images/ai-photoshoot-input-3e9da4e3f5e050cbe4f36788a62521de.png)
![Virtual try on J. Crew swimming suit](/assets/images/virtual-try-on-jcrew-swimming-suit-53c3b1768a8393476e21cf6489a9d5fc.webp)
More examples in the gallery - Swimming suit and Parka. See also the article walkthrough on virtual try-on.
Overview
Virtual try-on is the term for the process of generating images of a person wearing a specific garment or clothing item. This can be useful for e-commerce websites, fashion designers, or even for personal use.
To use Astria's virtual try-on feature, you need to create:
- A model of the person you want to dress up ideally through fine-tuning
- A faceid fine-tune of the garment you want to try on. The faceid class name should be
clothing
,shirt
orpants
. Class name is used to identify the garment in the image and remove the original model in case a reference image with a fashion model is used.
For prompting, use both the fine-tuned model of the person and the faceid reference of the garment. See below reference example prompt. Ideally you should try to describe the background/scene and type of cloth the person is wearing. In case a reference image to the garment includes a fashion model, a better alternative to this, could be to use the input image as a reference pose as well as img2img to bring some of the background features into the image. For example a swimming suit reference image might feature a background pool, while a reference parka might feature a snowy background.
It is recommended to use 3:4 aspect-ratio - 512x680
width x height for the input image.
<faceid:1227460:1> ohwx woman wearing a swimming suit, high resolution fashion photoshoot, UHD
num_images=4
negative_prompt=old, wrinkles, mole, blemish, scar, sad, severe, 3d, cg
seed=
steps=30
cfg_scale=
controlnet=pose
input_image_url=https://sdbooth2-production.s3.amazonaws.com/7cw0uwxugd4g22qobjlow6h5gbcv
mask_image_url=
denoising_strength=0.9
controlnet_conditioning_scale=0.4
controlnet_txt2img=false
super_resolution=true
inpaint_faces=true
face_correct=false
film_grain=false
face_swap=true
hires_fix=true
backend_version=1
prompt_expansion=undefined
ar=1:1
scheduler=dpm++sde_karras
color_grading=
use_lpw=true
w=
h=