Repository for performing feature matching using the DeDoDev2 model in PyTorch
Installation
Option1: pip installation
1
2
3
| git clone https://github.com/ibaiGorordo/dedodev2-pytorch-inference.git
cd dedodev2-pytorch-inference
pip install -r requirements.txt
|
Option2: uv installation
Recommended but requires uv to be installed (https://docs.astral.sh/uv/getting-started/installation/)
1
2
3
| git clone https://github.com/ibaiGorordo/dedodev2-pytorch-inference.git
cd dedodev2-pytorch-inference
uv sync
|
- Additionally, to activate the uv environment:
source .venv/bin/activate in macOS and Linux or .venv\Scripts\activate in Windows
Examples:
Image keypoint detection example
1
| python example_image_keypoints.py
|
Image pair matching example
1
| python example_image_pair_matching.py
|
Webcam feature tracking example
1
| python example_webcam_feature_tracking.py
|
Video feature tracking example (https://youtu.be/cDHNvORTRjo?si=aMZt5hiCrqWf3lhQ)
1
| python example_video_feature_tracking.py
|
License
The code is taken from the official DeDoDe repository which is distributed under the MIT License. See LICENSE for more information.
References:
