Repository files navigation

Metric depth estimation determines real-world distances to objects in a scene from images. This repo provides a web UI that allows users to estimate metric depth and visualize depth maps by easily uploading images using the Depth Pro model through a simple Gradio UI interface.

1. Create and activate a virtual environment:

   conda create -n depth-pro -y python=3.9conda activate depth-pro

2. Clone this repo and install the required library as follows:

   git clone https://github.com/spacewalk01/depth-pro-webui.gitcd depth-pro-webuipip install -r requirements.txt

3. Download the model:

   Download the model checkpointdepth_pro.pt and place it in thecheckpoints folder.

1. Launch the Gradio interface:

   python main.py

2. Open the provided local URL in your web browser to access the interface.

1. Upload an Image: Use the image uploader to select an image for depth estimation.
2. Adjust Options:
   • Auto Rotate: Enable or disable auto-rotation of the image.
   • Remove Alpha: Enable or disable the removal of the alpha channel.
3. View Results:
   • The depth map will be displayed on the interface.
   • The focal length in pixels will also be shown.

This project is licensed under the Apple License. See theLICENSE file for details.

• Gradio for providing an easy interface for machine learning models.
• Depth Pro for enabling sharp monocular metric depth estimation in less than a second.