Let's learn AI videos

We will discover text-to-video tools. After short research I propose to dive into `cogvideo`.

What we will to:

1. Read paper:
https://paperswithcode.com/paper/cogvideox-text-to-video-diffusion-models-with

2. Check repository
https://github.com/thudm/cogvideo

3. Reproduced demo from page locally:
https://huggingface.co/spaces/THUDM/CogVideoX-5B-Space

There is colab that we will study in practical part:
https://colab.research.google.com/drive/1pCe5s0bC_xuXbBlpvIH1z0kfdTLQPzCS?usp=sharing#scrollTo=3bBSgGvkz6y4

Taking participation in this event you:
- make new friends
- acquire skill of video generation by AI

Requirements:
- laptop with installed python, graphic card drivers
- preferred Linux / Mac, but we will help you with other systems too
- basic knowledge of python

So goal of this meeting is to acquire knowledge how cog video works from theoretical perspective and learn how to run it locally to generate videos from text.

`2024/9/19`: Cog Video have open-sourced the CogVideoX series image-to-video model CogVideoX-5B-I2V. This model can take an image as a background input and generate a video combined with prompt words, offering greater controllability. With this, the CogVideoX series models now support three tasks: text-to-video generation, video continuation, and image-to-video generation.

So it is hot 🔥🔥 and fresh 🧊🧊.

If you didn't had any experience with AI, don't worry. We will start from paper and will try to understand how it works.