2D-3D Reconstruction Project

Overview

This project aims to perform 2D to 3D reconstruction using computer vision techniques. It involves various steps such as camera calibration, feature extraction, sparse reconstruction, and bundle adjustment.

Files

• _2D_3D.py: This file contains the code for camera calibration using chessboard images.
• Feature_Extraction.py: This file is responsible for extracting features from object images.
• Sparse_Reconstruction.py: This file performs sparse reconstruction to generate 3D points.
• BundleAdjustment.py: This file optimizes the camera poses and 3D points using bundle adjustment.

Dependencies

• OpenCV
• NumPy
• OpenSfM

How to Run

1. Place your chessboard images in the Calibration Images folder.
2. Place your object images in the Object Images folder.
3. Run _2D_3D.py for camera calibration.
4. Run Feature_Extraction.py to extract features from object images.
5. Run Sparse_Reconstruction.py for sparse reconstruction.
6. Run BundleAdjustment.py for bundle adjustment.

License

This project is open-source and available under the MIT License.

Contributing

Feel free to contribute to this project by opening issues or submitting pull requests.