AK
RT @mervenoyann: Zero shot visual question answering with Frozen LLMs lets you answer questions based on images 👉🏼 https://t.co/oEC4acJ7ix https://t.co/B6oXY9PwM9
tweet

AK
RT @mervenoyann: Demo made for paper Text-Only Training for Image Captioning using Noise-Injected CLIP, made by @johko990 https://t.co/ilbQ8KjFRl
Link to paper: https://t.co/6oegBCIBgF https://t.co/7xLtohe5GC
tweet

AK
RT @mervenoyann: A demo made for the latest version of YOLO (v8) released by ultralytics 👀
Link to demo: https://t.co/APWXsxm6BT (@kadirnar_ai)
Link to GitHub repository: https://t.co/LQw0dNfzDU https://t.co/EEMVGRHCTI
tweet

AK
RT @mervenoyann: Realistic one-shot mesh-based head avatars (ROME) is a demo that lets you create an avatar both in video and image https://t.co/kcqQN2sHNq
it's based on a paper by Samsung https://t.co/8duTIYzda2 https://t.co/wyF3tY0iZQ
tweet

Carlos E. Perez
It's high time to make Fridays also part of the weekend! https://t.co/Zy4oXkva34
tweet

KaliYuga
RT @Morph_VGart: Whipped up a new animation. Tried to make it a gif but I guess I went too hard on it and the file go too big 😂

What would you title this? https://t.co/nyGz9YzTgm
tweet

Bojan Tunguz
What if all these LLMs turn out to be just cake?
tweet

KaliYuga
RT @nvnot_: gm, they grow up so fast https://t.co/hZboUMyISs
tweet

UnknownCollector - new 1/1s live 🔥
UnknownCollection

“Fading Away” by @itshamadsworld https://t.co/xzXVlxy7il
tweet

CoffeeVectors
RT @MrAI3D: These are not photographs, these are synths.
Creatures that (probably) do not exist.
They are just made using AI.

We are entering some extremely froggy times ahead🐸

#AIart #imgsynth #aiartist #midjourney #nijijourney https://t.co/OThYKfYX74
tweet

Altryne - targum.video
Asking for a friend (it's me, I'm the friend)
Did anyone train a diffusion model on like ALL THE ICONS ever so we can.. generate icons?

There's tons of amazing free icon sets out there, but the problem for non designers is, how to create custom ones.

(pictured heroicons) https://t.co/THPCzdh0QC
tweet