First Order Motion Model 中文展示版

First Order Motion Model 中文展示版

Makes an animated video from a source picture and a driving video, plus a side by side comparison, in a Chinese Gradio demo of First Order Motion Model. It suits learners and developers exploring motion transfer.

Made by n97141506 on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: This Space is a Chinese language demo of First Order Motion Model, built with Gradio. You upload a source picture and a driving video, and it produces an animated video from the two. It also makes a left and right comparison video and shows a run log on the page.

Who it suits: Chinese speaking learners and developers who want to try a motion transfer model through a simple web screen.

Industry | Who uses it | How they use it | Example
Education | Computer vision students | See how a motion model animates a still picture | A classroom demo with a short clip
Research | ML researchers | Compare the vox, vox-adv, taichi and fashion checkpoints | The same inputs on two checkpoints
Software | Developers | Read how a Gradio screen wraps the model | Opening app.py
Content | Demo presenters | Show motion transfer to an audience | A brief presentation clip

If you need long, high quality video output, you can skip it, since the free hardware is slow.

Where to use it:

• A web browser, on the Space's page at Hugging Face
• Python 3.10 on your own Windows, macOS or Linux computer, using a downloaded copy

Use case | Who | What they get back
Animating a picture | Learners | An animated video made from a source image
Side by side check | Researchers | A comparison video with the two shown together
Watching progress | Developers | A run log shown on the page

How to set it up:

• Python 3.10
• Python packages from requirements.txt, with Gradio 5.49.1
• No account or key is named

1. Install the Hugging Face hub package.
2. Download the Space with the hf command.
3. Move into the downloaded folder.
4. Install Gradio 5.49.1, then the requirements.
5. Start app.py.

pip install -U huggingface_hub
hf download n97141506/machine_learning --repo-type space --local-dir machine_learning
cd machine_learning
pip install gradio==5.49.1
pip install -r requirements.txt
python app.py

How to use it: Upload a source image and a driving video, choose a checkpoint such as vox, vox-adv, taichi or fashion, then run it. The page gives back the animated video, the comparison video and a log.

Tool | What it does
Source image | The picture you upload as the starting point
Driving video | The video you upload to drive the animation
Checkpoint | Your choice of vox, vox-adv, taichi or fashion
Max frames | A cap on the frames processed, where 0 tries the whole video

Example: You upload a portrait as the source image and a clip of about two seconds as the driving video. You keep the vox checkpoint, set max frames between 20 and 50 and run it. You get an animated video and a side by side comparison video.

For newcomers and experts: Start with the vox checkpoint, a clip of 1 to 3 seconds and max frames between 20 and 50. These small settings keep the wait short on the free hardware.

The Space defaults to free cpu-basic hardware, where video inference is slow. Setting max frames to 0 tries to process the entire clip, which may take a long time. The source names four checkpoints but does not describe how they differ.

You upload an image and a driving video
You pick a checkpoint and frame limit
The model builds the animated video
A comparison video and log are shown

Good to know:

• Runs on free cpu-basic hardware by default, so video work is slow
• Max frames set to 0 may need a long wait on that hardware
• License: mit
• Cost to run: none named, and the Space defaults to free hardware
• Data it reaches: the image and video you upload
• Not for: long clips on the free hardware, where 1 to 3 seconds is suggested

On Hugging Face, 13 November 2026

• Likes: 1
• Built with: gradio
• License: mit

Links

• Space: https://huggingface.co/spaces/n97141506/machine_learning
• Files: https://huggingface.co/spaces/n97141506/machine_learning/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.