Experimenting with Stable Video Diffusion - Moving Playground V2 Images

I've been experimenting with Stable Video Diffusion. I created 150 images using Playground V2 (1024x576) and then generated two videos from each image using Stable Video.

I then upscaled and converted these videos to 30 frames per second using Topaz Video AI. This clip is a selection from those videos.

While it's not perfect, I find it intriguing and believe it has the potential to spark some interesting ideas.

What do you think?

----

  / quick_eyed_sky   (to support, get prompts, ideas, and images)

My Colabs:

• Animagine XL 2.0: https://colab.research.google.com/dri...

• SDXL-DPO: https://colab.research.google.com/dri...

• Dreamlike Photoreal 2.0: https://colab.research.google.com/dri...

• Stable Diffusion SDXL + Kandinsky 3 + Playground V2: https://colab.research.google.com/dri...

• PlayGround V2: https://colab.research.google.com/dri...

• Stable Video Diffusion: https://colab.research.google.com/dri...

• AudioCraft MusicGen: https://colab.research.google.com/dri...

• QES MusicGEN: https://colab.research.google.com/dri...

• GFPGAN Face Restoration: https://colab.research.google.com/dri...

• Image-to-Text-to-Image: https://colab.research.google.com/dri...

• Kandinsky 2.1 + Symmetry + Dynamic Prompting + CSV + Upscaling: https://colab.research.google.com/dri...

• Kandinsky 2.1 + Automatic Outpainting: https://colab.research.google.com/dri...

• Deforum 7 + Dynamic Prompting: https://colab.research.google.com/dri...

• Kandinsky 3: https://colab.research.google.com/dri...



Playlist of AI tutorials:    • Tutorials for Disco Diffusion (and other AIs)   if you want to make your own images.

Some of my music you hear on my clips:   / user-894673824