Depth img2img - GTA V The Official Trailer - Weirder Version

Made using AUTOMATIC1111 Stable Diffusion in batch img2img mode using the depth model (512-depth-ema.ckpt). This is the first animation I've ever made using img2img. With this depth model, it's pretty nice. I did upscale and upsample to 60fps in post, and added a light touch of flicker control (temporal blur).

The prompt was "detailed colored pencil by r crumb and alex grey, psychedelic, 8k, octane render". I used 30 steps Euler_a, cfg_scale 30, denoising strength 0.5.

I'm currently in the middle of adding my Deforum mods to the AUTOMATIC1111 webui version. After that, I think I could remake this video using my optical flow as well!

#aiart #aiartist #deforum #stablediffusion

🌲⭐🌃
🤜 Forest Star Walz
👉 #reallybigname
https://www.youtube.com/reallybigname
https://www.facebook.com/reallybigname
https://www.instagram.com/reallybigname
https://www.pinterest.com/reallybigname
https://www.twitter.com/reallybigname
https://www.snapchat.com/add/reallybigname
https://www.soundcloud.com/forest-star-walz