The Aim is to reconstruct a 3D structure, given a sequence of images.
- first step is to compute the trjectory from sequence of images using Visual Odometry pipeline.
- pointcloud is then triangulated from two images. Here calculated the 3d points are the detected feature point from the pair of images which were used to estimate the trajectory.
- the pointcloud is the associated and can be visualized in meshlab.
git clone https://github.com/sushlokshah/structure_from_motion.git
cd structure_from_motion
#Edit info.yaml file
#dataset_location: [image_sequence_path]
#k: [calibration matrix]
python3 sfm.py
#Result is obtained in output.ply at folder specified, which can be viewed using MeshLab, Open3D, or any other related software or library to view point clouds.
Reconstructed structure | Trajectory |
---|---|