Learn how to create stunning AI avatars using Stable Diffusion and DreamBooth for free on Google Colab. Follow step-by-step instructions to train an open-source image generation model with your own portrait images, ensuring 100% privacy without relying on third-party apps. Discover how to prepare and upload training images, set optimal parameters, and generate a wide variety of personalized avatars in different styles. Explore techniques for fine-tuning prompts, adjusting settings, and batch-generating hundreds of unique images. Master the process of transforming selfies into professional-quality AI art, from initial setup to cherry-picking the best results, all without requiring technical expertise or a powerful computer.
Transform Your Selfie into a Stunning AI Avatar with Stable Diffusion - Better Than Lensa
Software Engineering Courses - SE Courses via YouTube
Overview
Syllabus
Introduction and the content
How to check you have enough Google drive space
Starting to prepare Google Colab notebook for Stablediffusion Dreambooth model training
Register, login and generate Hugging Face token for training of the AI model
Continue training setup of the Stable Diffusion Dreambooth
Setting the settings of Stable Diffusion Dreambooth model
Providing our own photos to train the model to teach our own face image
How to install Paint NET for image cropping
How to crop and prepare your training images by using Paint NET
How bulk image crop online with birme net website
How to quickly check all images dimensions in a folder by using sort options width, height in detailed folder view
How to upload training images to Google Colab
Starting training with last settings / options
Optimal parameters for training of Stable Diffusion Dreambooth model AI
Training of the AI model done
Exiting application/notebook completely and starting again to show how to use trained model
Starting to generate AI / Lensa app magic avatars
First avatar is generated and displayed
Explanation of the positive / inference prompt input to generate avatar images
Explanation of the negative prompt input to avoid such images, guidance_scale & num_inference_steps parameter of Stable Diffusion
New avatars with different guidance_scale parameter
How to generate colored portrait avatars / profile images
Showcasing different styles of generated avatars
Debated usage of artist styles
Continue to generate more artworks
Prompt for asian style artworks like korean, or japan or anime
Long hair, different eye color version of me
Adding armor keyword
Further tuning of input prompt
Cherry picking the results
How to generate hundreds or thousands of magic avatars as a batch
Clip of ~200 Stable Diffusion AI generated avatars
Ending talk and discussion
Taught by
Software Engineering Courses - SE Courses
Related Courses
-
Put Yourself Into a Universe of Art - Using Your Images in Dreambooth Stable Diffusion
-
Stable Diffusion Google Colab - Continue, Directory, Transfer, Clone, Custom Models, CKPT SafeTensors
-
Stable Diffusion Dreambooth Made Easy - Clone Yourself in AI Art
-
AI Avatars with Dreambooth Alternative LORA Based Stable Diffusion Fine-Tuning on Free Colab
-
How to Make AI Art - DreamBooth and Stable Diffusion Tutorial with Prompting Tips and Demo
-
Ultimate Dreambooth Guide for AI Art Creation
