Hello everyone. I am Dr. Furkan Gözükara. PhD Computer Engineer. SECourses is a dedicated YouTube channel for the following topics : Tech, AI, News, Science, Robotics, Singularity, ComfyUI, SwarmUI, ML, Artificial Intelligence, Humanoid Robots, Wan 2.2, FLUX, Krea, Qwen Image, VLMs, Stable Diffusion
I'm launching my site probably tomorrow along with my socials im just wrapping up its why im up late. The project I felt like doing earlier was I trained my subject but i captioned him as a woman and I used women for Class images because I wanted to use loras for womens clothing and to make it more versatile. And I was right.
I recently discovered Decorify (https://decorify.app/) and tested it with a living room image. Unlike my attempts with ControlNet, which maintained the furniture's structure, Decorify generated images with varied styles while preserving the room shape and rearranging the furniture in different shapes, like the diverse coffee table designs in each image. What might be their approach, and how does it differ from ControlNet?
So i had a chance to go through the Zero to Hero Stable Diffusion YouTube (great video btw). Althought our versions are slightly different, I was able to identify and set all the settings the way you did. Unfortunately, when I go to train it, it's just generating black pictures. So... I might be missing something. Any thoughts?