AI PAINTR
  • Home
  • Gallery
  • API
  • Resources
  • Pricing
  • Login
  • Blog
GET IN TO A

ai paintr

​STATE OF MIND

How To Generate Images

Background​


AI Paintr uses cutting edge AI techniques to train a model that can generate great looking images with just a text prompt. All you need to provide is some reference images. The reference images will be used to train a machine learning model that will learn the object of the reference images as a unique concept. Then you can use the model to generate new images for this unique concept.

For enquiries regarding image generation, collaboration and any service packages [e-commerce (product catalog), photography (maternity, new born, etc.)] contact us support@aipaintr.com

Steps to Train Model


1. Collect 10 to 30 images of yourself, your loved ones or an object. 
2. Resize the images to 512x512 size on birme.net. Download as a zip file. 
3. Go to https://aipaintr.com. Click on the app link. 
4. Create an account, add some credits.
5. Go to the “Train Model” link. 
6. Add a job name, choose a class, upload images, select a number of steps (preferably 3000). 
7. Wait for the training to complete
8. Once the model is ready. Go to the “Generate Images” link. 
​9. Generate cool images. Enjoy!

To skip resizing or format conversion to jpg/jpeg - you can click images in square mode & switch off HE photo format.

Original Images Specifications


Images Count    minimum 10 or maximum 50. To avoid overtraining and quicker results, can use a smaller count.

Concentration     selfies or solos, focusing on facial features with shoulders, whole body. Works for pets, couples, babies as well. 

Diversification     different backgrounds, poses, setups, clothing, facial expressions, gestures, props, lighting. Variation is better!
​

True Negation     setup repetition, shadows, glasses or hats (masks eyes, hairs - unless you prefer it that way), blurred or out of focus.

join us, let your imagination take control, it’s the beginning


Al Paintr © 2022
all rights reserved per federal & CA laws

FAQs
License and Credits:
(!) https://huggingface.co/spaces/CompVis/stable-diffusion-license
(2) https://github.com/CompVis/stable-diffusion
​(3) https://dreambooth.github.io/

Terms of Use

Content Policy
Privacy

reddit

​/u/aipaintr/

tiktok

@aipaintr
  • Home
  • Gallery
  • API
  • Resources
  • Pricing
  • Login
  • Blog