DeepDINO is a supervised depth estimation model that leverages the DINOv2 backbone for rich visual representations. DINO, which stands for Self-Supervised Vision Transformers, is introduced in the paper Learning Robust Visual Features without Supervision.
Clone the repository and install the necessary requirements:
git clone https://github.com/Tottowich/DeepDINO-DepthEstimation.git
cd DeepDINO-DepthEstimation
pip install -r requirements.txt
pip install git+https://github.com/facebookresearch/dinov2
Live videos of the model in action can be found in the videos folder. These are streams of the model running on a NVIDIA 3060 ti GPU on 720x1280 resolution.
We welcome contributions from the community! If you're interested in enhancing DeepDINO, please follow our Contributing Guidelines and Code of Conduct.
For documenting Python code, please adhere to the Totto-style docstring guidelines. Detailed guidelines can be found here.
[// License Information]