#computer-vision #algorithm #photogrammetry #camera #linear #transform #3d

no-std dlt

DLT (direct linear transform) algorithm for camera calibration

15 releases (breaking)

0.12.0 Aug 28, 2023
0.11.0 May 2, 2022
0.10.0 Apr 6, 2022
0.9.1 Sep 27, 2021
0.2.2 Mar 22, 2020

#609 in Algorithms

Download history 84/week @ 2023-11-20 93/week @ 2023-11-27 104/week @ 2023-12-04 139/week @ 2023-12-11 106/week @ 2023-12-18 21/week @ 2023-12-25 22/week @ 2024-01-01 21/week @ 2024-01-08 82/week @ 2024-01-15 260/week @ 2024-01-22 304/week @ 2024-01-29 305/week @ 2024-02-05 87/week @ 2024-02-12 136/week @ 2024-02-19 242/week @ 2024-02-26 240/week @ 2024-03-04

755 downloads per month

MIT/Apache

23KB
293 lines

Crate dlt for the Rust language

Crates.io Documentation Crate License Dependency status build

DLT (direct linear transform) algorithm for camera calibration

This is typically used for calibrating cameras and requires a minimum of 6 corresponding pairs of 2D and 3D locations.

Testing

Unit tests

To run the unit tests:

cargo test

Test for no_std

Since the thumbv7em-none-eabihf target does not have std available, we can build for it to check that our crate does not inadvertently pull in std. The unit tests require std, so cannot be run on a no_std platform. The following will fail if a std dependency is present:

# install target with: "rustup target add thumbv7em-none-eabihf"
cargo build --no-default-features --target thumbv7em-none-eabihf

Example

use dlt::{dlt_corresponding, CorrespondingPoint};

let points: Vec<CorrespondingPoint<f64>> = vec![
    CorrespondingPoint {
        object_point: [-1., -2., -3.],
        image_point: [219.700, 39.400],
    },
    CorrespondingPoint {
        object_point: [0., 0., 0.],
        image_point: [320.000, 240.000],
    },
    CorrespondingPoint {
        object_point: [1., 2., 3.],
        image_point: [420.300, 440.600],
    },
    CorrespondingPoint {
        object_point: [1.1, 2.2, 3.3],
        image_point: [430.330, 460.660],
    },
    CorrespondingPoint {
        object_point: [4., 5., 6.],
        image_point: [720.600, 741.200],
    },
    CorrespondingPoint {
        object_point: [4.4, 5.5, 6.6],
        image_point: [760.660, 791.320],
    },
    CorrespondingPoint {
        object_point: [7., 8., 9.],
        image_point: [1020.900, 1041.800],
    },
    CorrespondingPoint {
        object_point: [7.7, 8.8, 9.9],
        image_point: [1090.990, 1121.980],
    },
];

let pmat = dlt_corresponding(&points, 1e-10).unwrap();
// could now call `cam_geom::Camera::from_perspective_matrix(&pmat)`

See also

You may also be interested in:

  • cam-geom - Rust crate with 3D camera models which can use the calibration data from DLT.
  • dlt-examples
    • Unpublished crate in the dlt repository which demonstrates usage with cam-geom library.

Regenerate README.md

The README.md file can be regenerated with:

cargo readme > README.md

Code of conduct

Anyone who interacts with this software in any space, including but not limited to this GitHub repository, must follow our code of conduct.

License

Licensed under either of these:

Dependencies

~3.5MB
~73K SLoC