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
Our Discord : https://discord.gg/HbqgGaZVmr. This is the video where you will learn how to use Google Colab for Stable Diffusion. If I have been of assistance to you and you would like to show your support for my work, please consider becoming a patron on https://www.patreon.com/SECourses
Playlist of Stable Diffusion Tutorials, #Automatic1111...
Thank you for answering! Do you think an outside 3rd party upscale is an option? Or is it that once the video is generated, it is what it is and it can’t be improved?
@StarBuffer it is very hard to get exact images of model you trained from this shoe, in case of lora. expecially when your shoe has many stylized detials. checkpoint may be better. Otherwise you might need 50 images for training. sd 1.5 training is very bad. I suggest you not to use.
@Dr. Furkan Gözükara my model trained on realistic was unable to get my face at all checkpoints. I didn't use Lion but 8bit Adam with the settings for 12GB Vram.
with 8bit adam and xformers my training took 2:55hs now with lion and without xformers it will take almost 7 hours with 20 images in the dataset i will try kaggle
I had to look for an example openpose video to use with my image, but the result was black, I think it takes a long time to process, but it also didn't give any error messages.