Is this how future Human-AI collaboration can look like? #ChatGPT even understands how to build weighted multi-prompts for #midjourney with some guidance. Even though it ignored my "ugly hands" negative prompt request. ChatGPT probably knows diffusion models can't draw hands 😅
DeepMind & UCL Deep Learning Lecture Series
12 lectures on various deep learning algorithms and techniques in natural language processing, vision, generative models, etc.
All lectures & presentations are amazing, free and you can take them at your pace.
https://t.co/QnXdvb56Tg
Version 0.7.2 of the @StableDiffusion@Photoshop plugin brings dramatically improved inpainting and CLIP guidance. Both are automatic. You can override the diffusion engine if you want, but by default, the plugin will use the best engine for the task.
https://t.co/gqFWpABQLY
Did you know that you could fine-tune your own Stable Diffusion model, host it online and have ~3.5s inference latency, all cost-free? Here's the step-by-step tutorial on how to do it: 🎨
Colab link to follow along: https://t.co/951XsQHDsN (h/t TheLastBen)
🧵 (takes around 2hrs)
🧑💻 video transparency tutorial 🧑💻
How to create and compress transparent videos for all browsers.
How to use them in WebGL (to compress even more).
How to avoid transparent videos.
https://t.co/GW4eXqixS7
https://t.co/WLGbXHj17f is an outstanding event! Very smart conceptually, beautiful designwise; very funny, but also serious education! Should be held once a week all around the globe, online and irl! 👏👏👏👏
After years of making content for free, I completed my first ever paid-for course over at https://t.co/YKcQaTxeJB ✨
It teaches front-end devs how to make Awwwards-type websites. Even if you've never even touched Threejs before.
RT appreciated! 😊
The sixth of 10 vans went to auction 🔥
As a reminder, the owner of this token will get a real graffiti on the train with his name, this wagon will travel all over Ukraine 🇺🇦
https://t.co/bPW1R8s12t
#nftart#nftcommunity#graffiti