Diffusion: Generating Images from Noise - Generative AI for Developers

5 Просмотры

  • Спасибо! Поделитесь с друзьями!

    Вам не понравилось видео. Спасибо за то что поделились своим мнением!

aii
Издатель
Diffusion: Generating Images from Noise - part of Generative AI for Developers.
How LLMs actually work - and how to build with them. Visual, developer-first.

In this lesson:
- Images emerge by removing noise step by step
- The prompt guides every single step
- Latent space makes it fast and practical
- Steps and guidance are your two main dials

Chapters:
0:00 Intro
0:05 Intro
0:22 How do you build an image from scratch?
0:39 One mechanism, many media
0:56 Analogy
1:15 Training idea: add noise
1:32 The model learns the reverse
1:48 Generation: start from nothing
2:04 Where does the prompt come in?
2:25 Guidance strength
2:44 Latent diffusion
3:08 The whole pipeline
3:30 Denoising, in miniature
3:51 The real thing (API)
4:14 What diffusion is NOT
4:35 Using diffusion well
4:58 Recap

Module: Seeing & Hearing (Multimodal)
Every command and code example in this video is executed and verified before publishing.
Категория
AI Картинки
Выполните вход или зарегистрируйтесь для отправки комментария.
Комментариев нет.