DALL-E
lastmod 2023-02-21created 2022-04-089 edits
DALL-E (stylized DALL·E) is an artificial intelligence program that creates images from textual descriptions, revealed by OpenAI on January 5, 2021. It uses a 12-billion parameter version of the GPT-3 Transformer model to interpret natural language inputs (such as "a green leather purse shaped like a pentagon" or "an isometric view of a sad capybara") and generate corresponding images. It can create images of realistic objects ("a stained glass window with an image of a blue strawberry") as well as objects that do not exist in reality ("a cube with the texture of a porcupine"). Its name is a portmanteau of WALL-E and Salvador Dalí.
DALL-E 2
- https://openai.com/dall-e-2/
- An AI system that generates realistic images from natural language descriptions
- Version 2, released 1 year and 2 months later, generates more realistic and accurate images at 4x the resolution
- Can combine concepts/attributes/styles
- an astronaut/teddy bear/bowl of soup, riding a horse/lounging at a resort/playing basketball, as a photorealistic/Andy Warhol style/pencil drawing image
See also
- OpenAI
- GPT-3
- Imagen
- Text-to-Image
- Stable Diffusion - open-source Text-To-Image similar to DALL-E
- DreamFusion - Text-to-3D using 2D Diffusion
- Make-A-Video - Text-to-Video Generation without Text-Video Data
- Creating Moe Illustrations Using AI Tutorial - a tutorial on generating moe illustrations with AI