Video yükleniyor...
Video Yüklenemedi
New way to navigate latent space. It preservers the underlying image structure and feels a bit like a powerful style-transfer that can be applied to anything. The trick is to...
313,838 görüntüleme • 1 yıl önce •via X (Twitter)
9 Yorum

selectively alter the embeddings in the decoder part of the diffusion process. The demo is powered by SDXL Turbo and is running in realtime. The MIDI controller is a great way of modifying variables in real time (see The prompts were...

"photo of a red brick house, blue sky" as base prompt, the new decoder embeddings were "coral", "moss", "fire", "ice", "sand", "rusty steel" and "cookie".

How could something like this be rigged up to an instrument. So when you hit certain notes or chords those particular frequencies facilitate image manipulation. I always thought it would be awesome to create art or visuals directly through an instrument or some form of sound interface.

absolutely! working on coupling this to real time audio generation, leveraging synesthesia & immersion

loving this! you got it. thank you!

Ah yes, the eight elements: fuzz, moss, fire, ice, marble, sand, grunge, and bread.

This is soooo cool! I could definitely see that in an art or science museum, where the visitor could play with it in real time. Nice concept!!

have done it with @lunarringart a while ago with good old VQGAN

it's very impressive, but that is an absolutely crazy way to turn knobs. are we sure the bottom is not also AI generated?
