https://github.com/henauxg/bitmaps_to_triangulation_data

https://github.com/henauxg/bitmaps_to_triangulation_data

Last synced: about 6 hours ago
JSON representation

• Host: GitHub
• URL: https://github.com/henauxg/bitmaps_to_triangulation_data
• Owner: Henauxg
• License: mit
• Created: 2024-07-01T13:08:33.000Z (5 months ago)
• Default Branch: main
• Last Pushed: 2024-07-20T10:58:00.000Z (4 months ago)
• Last Synced: 2024-10-15T11:52:41.422Z (30 days ago)
• Language: Rust
• Size: 24.4 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE-MIT

Awesome Lists containing this project

README

        
### Bitmaps to contour edges & vertices

This repository contains the tool used to produce the edges & vertices data for the "Bad Apple" example in https://github.com/Henauxg/ghx_constrained_delaunay.

See a video result here: https://www.youtube.com/watch?v=TrfDkD_PprQ

You can find serialized datasets produced by this tool in https://github.com/Henauxg/cdt_assets

### Usage

To reproduce:

- Take an input video

- Extract each frame of the video to a bitmap file, using [ffmpeg](https://www.ffmpeg.org/) (or any other similar tool):

  - `ffmpeg.exe -i 'video_file' .\frames\%05d.bmp`

- Update `FRAMES_PATH` and `FRAMES_TO_PROCESS_RANGE` in `main.rs`

- Run with `cargo run --release`

- Output frames data will be in `frames.msgpack`

### Process

This was made as a one-shot tool, code is not clean nor well organized.

For each greyscale bitmap image, it will:

-  convert it to a black & white monochrome image

-  detect the edges

-  find multiple paths from the edges

-  create domains from the paths

-  hierarchize the domains (inclusion)

-  find the paths/domains orientations

-  from the oritentations/hierarchy, serialize the paths using [msgpack-rust](https://github.com/3Hren/msgpack-rust) as a `Vec` of

    ```rust

    pub struct Frame {

        pub vertices: Vec<(i32, i32)>,

        pub edges: Vec<(usize, usize)>,

    }

    ```

### Notes

Although it was only used and tested on some "Bad Apple" videos (https://www.youtube.com/watch?v=FBqz4WzE1sc and https://www.youtube.com/watch?v=FtutLA63Cp8), the tool should still work on any greyscale video, in any resolution.

### License

All code in this repository is licensed under the MIT License ([LICENSE-MIT](LICENSE-MIT) or [http://opensource.org/licenses/MIT](http://opensource.org/licenses/MIT))