DA-2

DA-2

Runs the DA^2 depth model in a browser demo, the official demo of Depth Anything in Any Direction. For researchers and developers who want to try depth estimation without setting anything up.

Made by haodongli 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: DA-2 is the official demo of DA^2, a depth model whose title reads Depth Anything in Any Direction. The Space lets you try the model in a browser. A model and a dataset with the same name, haodongli/DA-2, are linked to it.

Who it suits: People who study or build on depth estimation and want to see this model run before reading its code.

Industry | Who uses it | How they use it | Example
Research | Computer vision researchers | Try the model on their own pictures | Run the demo and look at the depth result
Software | Developers adding depth to an app | Judge whether the model fits their need | Test a few sample inputs
Education | Students of computer vision | Learn what depth estimation produces | Open the demo and try one picture

You can skip it if you do not work with images or depth, since the Space does one job.

Where to use it:

• A web browser, on the Space's page at Hugging Face, with nothing to install
• Windows, macOS or Linux with Python, from a copy you download

Use case | Who | What they get back
Try the model | Researchers | A depth result for the picture you add
Check the setup | Developers | A working copy of the demo on your computer

How to set it up:

• Python on your computer
• The Python packages listed in requirements.txt
• The huggingface_hub package, which gives you the hf command

1. Install the huggingface_hub package.
2. Download the Space with the hf command.
3. Move into the downloaded folder.
4. Install the pinned Gradio version.
5. Install the packages from requirements.txt.
6. Start the app with Python.

pip install -U huggingface_hub
hf download haodongli/DA-2 --repo-type space --local-dir DA-2
cd DA-2
pip install gradio==5.49.1
pip install -r requirements.txt
python app.py

How to use it: Open the Space's page in your browser, or start your own copy with the command below. Then add a picture in the interface and read the result it returns.

python app.py

Example: You add a picture to the demo and run it. The app returns the depth result for that picture so you can judge how the model reads the scene.

For newcomers and experts: If you are new to depth estimation, open the page and try one picture you know well. Compare what the model returns with how near or far things are in the scene.

If you know the field, the Space is a Gradio front end for the DA^2 model. The repository holds app.py, a configs folder and a da2 package. The Space is set to GPU hardware at Hugging Face, so a local copy may need a capable graphics card for speed.

You add a picture in the Gradio interface
app.py passes it to the DA^2 model
The model estimates depth
The result shows on the page

Good to know:

• License: the Space lists apache-2
• Cost to run: none named
• Data it reaches: the pictures you add, plus the model and files in the Space
• Not for: anything the source does not describe, since the README is a single line about the official demo

On Hugging Face, 13 November 2026

• Likes: 49
• Built with: gradio
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/haodongli/DA-2
• Files: https://huggingface.co/spaces/haodongli/DA-2/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.