My gut feeling also told me to not chase hype but go where my curiosity leads me which is photorealistic AI renders, not avatars I did manage to set up my own GPU now and train my own models for that!
MULTIMODAL AI
-
Image Generation Quality Issues and Current Technology Limitations
By
–
Use better input pics, and even then many of pics are unusable and have artefacts. Reality of new tech
-
Stable Diffusion Fine-tuned on Thousands of Images
By
–
Stable Diffusion then finetuned on thousands pics
-
AI-Generated Person Looks Surprisingly Realistic
By
–
This person esp looks quite real to me, I'm kinda shocked
-
Training Custom Stable Diffusion Model with Personal Dataset
By
–
Pics of my first FULLY self trained Stable Diffusion (not Dreambooth) with my own data set These are just samples as it's not even fully done training Will train on men next! Then a combo model
-
Stable Diffusion-based Text-to-3D Model Demo Released
By
–
“Stable Diffusion for 3D” is coming
— AI Breakfast (@AiBreakfast) 21 décembre 2022
Here’s a Text-to-3D point map demo on @huggingface by @hahahahohohe https://t.co/6DPjXI2Hg2
“Mid century modern chair” prompt returned this. pic.twitter.com/qrNGINKg0N“Stable Diffusion for 3D” is coming Here’s a Text-to-3D point map demo on @huggingface by @hahahahohohe https://
huggingface.co/spaces/anzorq/
point-e_demo
… “Mid century modern chair” prompt returned this. -
Translation Glasses App Test: Real-Time French Subtitles Experience
By
–
Tested translation!
— Allie K. Miller (@alliekmiller) 20 décembre 2022
I’m wearing the glasses while the app is translating – it’s showing me the French subtitles as I speak.
Pros: fast, 9 languages, easy to use, easy to read in AR
Cons: have to specify “to and from” languages, lower accuracy when audio is hard (action movie) pic.twitter.com/6JWucXATktTested translation! I’m wearing the glasses while the app is translating – it’s showing me the French subtitles as I speak. Pros: fast, 9 languages, easy to use, easy to read in AR Cons: have to specify “to and from” languages, lower accuracy when audio is hard (action movie)
-
OpenAI Open-Sources Point-E: Fast 3D Point Cloud Generation
By
–
Just open-sourced Point·E — DALL·E but for 3D point clouds. Extremely fast to generate: https://t.co/fXrdnxi7Qw pic.twitter.com/PUKRZO9fum
— Greg Brockman (@gdb) 20 décembre 2022Just open-sourced Point·E — DALL·E but for 3D point clouds. Extremely fast to generate: https://
github.com/openai/point-e -
Flamingo AI Advances Computer Vision Beyond Past Predictions
By
–
10 yrs ago @karpathy wrote a blog post on the outlook of AI: https://
karpathy.github.io/2012/10/22/sta
te-of-computer-vision/
… in which he describes how difficult it would be for an AI to understand a given photo, concluding "we are very, very far and this depresses me."
Today, our Flamingo steps up to the challenge. -
Testing Flamingo AI’s Image Recognition Against Parental Challenge
By
–
The "Parents AI test". My dad just sent me a picture and told me: "there's no way AI can guess what this is". Challenge accepted! #Flamingo