Uniformer_video_demo

Uniformer_video_demo

A Gradio demo that labels short videos with the Andy1621/uniformer model, with three sample clips. It suits people who want to try video classification in a browser.

Made by Andy1621 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: Uniformer_video_demo is a Gradio app on Hugging Face Spaces that labels short videos with the Andy1621/uniformer model. The repository holds the model code, a Kinetics class index and three sample clips, hitting baseball, hoverboarding and yoga. It is built with Gradio 3.0.3.

Who it suits: It suits people who want to try video classification in a browser, and learners who want to see how a Uniformer model is wrapped in a Gradio page.

Industry | Who uses it | How they use it | Example
Machine learning | Students | Read uniformer.py and app.py to see how the model is called | Download the Space and open the two files
Computer vision | Researchers | Try the model on a quick sample clip | Click the yoga clip on the Space page
Education | Instructors | Show action recognition with ready samples | Run the Space in a lesson

Anyone who needs documented accuracy or a way to train the model can skip it, since the source gives none.

Where to use it:

• A web browser, straight from the Space page on Hugging Face
• Windows, macOS or Linux with Python, as a local copy

Use case | Who | What they get back
Label a video clip | Anyone curious | A predicted action label for the clip, shown in the browser
Study a model demo | Learners | Model code, label index and sample clips in one repository

How to set it up: Using it online needs nothing. A local copy needs these first.

• Python and pip
• The huggingface_hub package, to fetch the files
• Gradio at version 3.0.3, and the packages in requirements.txt

1. Install the Hugging Face Hub package.
2. Download the Space into a folder called uniformer_video_demo.
3. Move into that folder.
4. Install Gradio 3.0.3, then the packages from requirements.txt.
5. Start the app.

pip install -U huggingface_hub
hf download Andy1621/uniformer_video_demo --repo-type space --local-dir uniformer_video_demo
cd uniformer_video_demo
pip install gradio==3.0.3
pip install -r requirements.txt
python app.py

How to use it: Open the Space page on Hugging Face, then give it a video or pick a sample clip. For a local copy, run python app.py in the folder and open the page it starts.

Example: You pick the hoverboarding clip on the Space page. The app runs the Uniformer model on it and shows the action label it predicts for the clip.

For newcomers and experts: If you are new, click one of the three sample clips first. That shows the whole idea in one step, with nothing to install.

If you know the field, the model code sits in uniformer.py, frame handling in transforms.py and the label names in kinetics_class_index.py, so the demo labels actions from the Kinetics set. Gradio is pinned to 3.0.3, an early release.

You open the page and choose a clip
transforms.py prepares the video frames
uniformer.py predicts the action label

Good to know:

• The README gives no description, so settings and limits are clear only from the page and the code
• It is pinned to Gradio 3.0.3, so an older environment may suit it best
• License: mit
• Cost to run: none named
• Data it reaches: the videos you give it and the Andy1621/uniformer model it names

On Hugging Face, 13 November 2026

• Likes: 7
• Built with: gradio
• License: mit

Links

• Space: https://huggingface.co/spaces/Andy1621/uniformer_video_demo
• Files: https://huggingface.co/spaces/Andy1621/uniformer_video_demo/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.