Crate locus_core 

High-performance AprilTag and ArUco detection engine.

Locus is a research-oriented, memory-safe fiducial marker detector targeting low latency. It provides a performance-focused pipeline for robotics and computer vision, with strict zero-heap allocation in the detection hot-path.

Cargo features

• non_rectified — enables [BrownConradyModel] and [KannalaBrandtModel] alongside the corresponding pose::DistortionCoeffs variants. Off by default; opt in for cameras with significant lens distortion (OpenCV Brown-Conrady or Kannala-Brandt equidistant fisheye). See README.md for the full feature matrix.

Re-exports

pub use crate::board::AprilGridTopology;

pub use crate::board::BoardConfigError;

pub use crate::board::CharucoTopology;

pub use crate::camera::CameraModel;

pub use crate::camera::PinholeModel;

pub use crate::config::CornerRefinementMode;

pub use crate::config::DecodeMode;

pub use crate::config::DetectOptions;

pub use crate::config::DetectorConfig;

pub use crate::config::PoseEstimationMode;

pub use crate::config::QuadExtractionMode;

pub use crate::config::TagFamily;

pub use crate::detector::Detector;

pub use crate::detector::DetectorBuilder;

pub use crate::detector::FrameContext;

pub use crate::detector::LocusEngine;

pub use crate::error::ConfigError;

pub use crate::error::DetectorError;

pub use crate::image::ImageView;

pub use crate::pose::CameraIntrinsics;

pub use crate::pose::DistortionCoeffs;

pub use crate::pose::Pose;

Modules

board

Board-level configuration and layout utilities. Board-level configuration and layout utilities.

camera

Zero-cost camera distortion model trait and implementations. Zero-cost camera distortion model abstractions.

charuco

ChAruco board saddle-point extraction and pose estimation. ChAruco board saddle-point extraction and board pose estimation.

config

Configuration types for the detector pipeline. Configuration types for the detector pipeline.

detector

The primary public API for the detector.

error

Error types for the detection pipeline. Error types for the detection pipeline.

image

Image buffer abstractions. Stride-aware image view for zero-copy ingestion.

pose

3D Pose Estimation (PnP). 3D Pose Estimation (PnP) for fiducial markers.

Structs

Detection

A single tag detection result.

Point

A 2D point with f64 precision.

TelemetryPayload

Metadata for zero-copy telemetry extraction.

Functions

core_info

Returns version and build information for the core library.