vision-calibration

High-level entry crate and facade for the calibration-rs toolbox.

This is the recommended crate for most users. It re-exports all sub-crates through a unified API.

Features

• Session API: Structured calibration workflows with step functions, state tracking, and JSON checkpointing
• 6 workflows: Planar intrinsics, Scheimpflug intrinsics, single-camera hand-eye, rig extrinsics, rig hand-eye, laserline device
• Prelude module: Minimal imports for planar "hello world" calibration
• Foundation access: Direct access to core types, linear solvers, and optimization when needed

Quick Start

Add to your Cargo.toml:

[dependencies]
vision-calibration = { git = "https://github.com/VitalyVorobyev/calibration-rs" }

Planar Intrinsics Calibration

use vision_calibration::prelude::*;
use vision_calibration::planar_intrinsics::{step_init, step_optimize};

let mut session = CalibrationSession::<PlanarIntrinsicsProblem>::new();
# let dataset: PlanarDataset = unimplemented!();
session.set_input(dataset)?;

step_init(&mut session, None)?;
step_optimize(&mut session, None)?;

let result = session.export()?;

Single-Camera Hand-Eye Calibration

use vision_calibration::session::CalibrationSession;
use vision_calibration::single_cam_handeye::{
    SingleCamHandeyeProblem,
    step_intrinsics_init, step_intrinsics_optimize,
    step_handeye_init, step_handeye_optimize,
};

let mut session = CalibrationSession::<SingleCamHandeyeProblem>::new();
# let input = unimplemented!();
session.set_input(input)?;

step_intrinsics_init(&mut session, None)?;
step_intrinsics_optimize(&mut session, None)?;
step_handeye_init(&mut session, None)?;
step_handeye_optimize(&mut session, None)?;

let result = session.export()?;

Scheimpflug Intrinsics Calibration

use vision_calibration::core::PlanarDataset;
use vision_calibration::session::CalibrationSession;
use vision_calibration::scheimpflug_intrinsics::{
    ScheimpflugIntrinsicsConfig, ScheimpflugIntrinsicsProblem, run_calibration,
};

# fn main() -> anyhow::Result<()> {
# let dataset: PlanarDataset = unimplemented!();
let mut session = CalibrationSession::<ScheimpflugIntrinsicsProblem>::new();
session.set_input(dataset)?;

let config = ScheimpflugIntrinsicsConfig::default();
run_calibration(&mut session, Some(config))?;
let result = session.export()?;
println!("mean reprojection error: {:.4}", result.mean_reproj_error);
# Ok(())
# }

Using the Prelude

use vision_calibration::prelude::*;

// Minimal hello-world imports:
// CalibrationSession + planar problem + planar runner + core planar types.

Available Problem Types

Each problem type also provides a run_calibration convenience function that runs all steps.

Module Organization

When to Use This Crate vs. Sub-Crates

Examples

See examples/ directory:

See Also

• vision-calibration-core: Core primitives
• vision-calibration-linear: Linear solvers
• vision-calibration-optim: Non-linear optimization
• vision-calibration-pipeline: Pipelines and session API