Sports2d

Sports2d

Works out a person's 2D body pose and joint angles from a video file or a webcam feed, in a Gradio page. It suits coaches, movement researchers and anyone studying how a body moves.

Made by DavidPagnon 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: Sports2d is a Gradio app that reads a video or a live webcam feed and estimates where a person's body sits in each frame, together with the angles of the joints. It saves you from marking up footage by hand. The Space runs on Hugging Face's free CPU hardware and ships with four sample clips.

Who it suits: It suits people who study human movement and want quick numbers from ordinary video, without lab equipment.

Industry | Who uses it | How they use it | Example
Sports coaching | Coaches and athletes | Check body positions and joint angles in a training clip | A coach loads a jump clip and looks at the angles at takeoff.
Healthcare | Physiotherapists | Look at how a person walks on film | A therapist runs a walking clip to see the angles of the legs.
Research | Movement scientists | Get pose estimates from a camera without special hardware | A student tests the app on a clip before building a larger study.
Software | Developers | Read a small pose app built on Gradio | A developer starts app.py locally to see how it is wired.

If you need exact clinical measurements or 3D data, skip it, since the app works in two dimensions.

Where to use it:

• A web browser, through the Space's page on Hugging Face
• A local copy on Windows, macOS or Linux, using Python

Use case | Who | What they get back
Analyze a video | Coaches and researchers | Estimated 2D pose and joint angles for the clip
Try the webcam | Anyone | The same estimates from a live camera feed
Test with sample clips | Newcomers | Results for the bundled jump, walking and hurdle videos

How to set it up:

• Python and pip on Windows, macOS or Linux
• The packages listed in requirements.txt

1. Install the huggingface_hub package.
2. Download the Space into a local folder named sports2d.
3. Move into that folder.
4. Install gradio at version 6.14.0.
5. Install the packages listed in requirements.txt.
6. Start the app.

pip install -U huggingface_hub
hf download DavidPagnon/sports2d --repo-type space --local-dir sports2d
cd sports2d
pip install gradio==6.14.0
pip install -r requirements.txt
python app.py

How to use it: Open the Space's page, or start the app locally with the command below. Then give it a video file or switch to your webcam and run the analysis.

python app.py

Tool | What it does
app.py | The main file, which starts the Gradio page
Sample videos | Four short clips in the repository to try the app on

Example: You pick the hurdle_slowmo.mp4 sample clip and run the analysis. The app returns the estimated body pose for the clip, along with the joint angles it worked out.

For newcomers and experts: If you are new, open the Space's page and use one of the four sample clips. That shows what the app produces before you bring your own footage.

If you know the field, this is 2D pose estimation, so angles are measured in the image plane and depend on camera position. Running on a free CPU Space means long clips take time.

Choose a video or the webcam
The app finds the body in each frame
Joint angles are computed
Results appear in the page

Good to know:

• It works in 2D, so camera angle affects the numbers
• The source does not describe an export format for results
• License: bsd-3-clause
• Cost to run: none named
• Data it reaches: the video files you give it and, if you choose it, your webcam

On Hugging Face, 13 November 2026

• Likes: 12
• Built with: gradio
• License: bsd-3-clause

Links

• Space: https://huggingface.co/spaces/DavidPagnon/sports2d
• Files: https://huggingface.co/spaces/DavidPagnon/sports2d/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.