Video Tour Generation from PLY Point Cloud

This repository provides a system for generating a video tour from a given .ply point cloud file. It includes several scripts for processing the point cloud, planning a camera path, and rendering the tour.

Before running any scripts, make sure all libraries in requirements.txt are installed in your environment.

---

1. explorator.py

• Loads a .ply file.
• Detects the floor plane.
• Automatically fixes axis orientation.
• Builds an occupancy grid.
• Saves the free points to a JSON file named free_points.json.

To run:

py explorator.py

---

2. path_planner.py

• Loads free_points.json.
• Finds starting points.
• Generates a smooth path using a simple algorithm.
• Saves the path points into smooth_path.json.

To run:

py path_planner.py

---

3. index.html (renderer)

• Renders the scene from the .ply points.
• Sets the camera on every point along the path.
• Provides smooth transitions between points.
• Records the video of the tour.

To run the renderer server:

py -m http.server 8000

To open and start the video recording:

• Open in browser: http://localhost:8000
• When the recording is finished, the video will be downloaded in .webm format to your local storage (a notification will appear).

---

Additional Notes

• All .ply were converted using superspl.at but was not loaded because of size.
• For path planning, random steps in the range (0.01, 0.05) were used for every new point.
• Video are saved in webm format, but converted manually for github.*