Fascinating & kinda cool. On the other hand, this illustrates the clear & present danger of not getting specific, detailed, bulletproof AI contract language. This is what happens when studios want to use AI without actors @WGAWest@sagaftra
Well, @runwayml's extended clip generation couldn't have come at a better time! 🎬
8-12 second generated clips do so much for atmosphere building.
Here's a little feeler of the world that I'm creating for the full-length version of my AI generated short, Doctrine.
Finally, I figured out how to produce incredible cinematic shots in #midjourneyV4. The new ability to use a 16:9 ratio helps a lot too. Here's a pirate adventure, but directed by Sergio Leone.
@tailwindcss is amazing. But remembering every class and deciphering other people's code can be a headache.
So I built "UnwindCSS," an AI tool that translates classes to CSS.
Also, supports natural language ➡ utility classes.
https://t.co/2WkedoT9Tt
Accurate Pose Estimation Via WiFi Signals
"our model can estimate the dense pose of multiple subjects, with comparable performance to image-based approaches, by utilizing WiFi signals as the only input"
arxiv: https://t.co/sV5X1ZQLWS
OpenAI just dropped a prototype of 3D DALLE (called “Point-E”) 👀. It isn’t as good as Google’s DreamFusion, but blazing fast! Like ~600x faster to generate 😮. 2D DALLE has already turned the creative world upside down. How will 3D DALLE disrupt games, VR, metaverse, …? 🤯
Scalable Diffusion Models with Transformers
abs: https://t.co/RlOulZLZ1U
largest DiT-XL/2 models outperform all prior diffusion models on the class conditional ImageNet 512×512 and 256×256 benchmarks, achieving a state-of-the-art FID of 2.27 on the latter