PoseLib/PoseLib

Minimal solvers for calibrated camera pose estimation

What it solves

PoseLib is a library of minimal solvers for camera pose estimation. It addresses the problem of determining a camera's position and orientation (pose) based on different types of correspondences, such as point-to-point, point-to-line, and line-to-line mappings between 2D image data and 3D world data.

How it works

The library provides a collection of specialized solvers that calculate the pose from a minimal set of correspondences. It uses a consistent calling interface and minimizes external dependencies (relying primarily on Eigen).

To handle noise and outliers in real-world data, it implements robust estimators based on LO-RANSAC. These estimators normalize data, generate hypotheses using the minimal solvers, and then apply non-linear refinement (minimizing reprojection or Sampson error) to produce a final, accurate pose.

It supports various camera models (including Pinhole, Radial, and OpenCV Fisheye) and provides Python bindings for easy integration.

Who it’s for

This library is designed for developers and researchers working in computer vision, SLAM (Simultaneous Localization and Mapping), and 3D reconstruction who need fast, robust, and stand-alone solvers for camera pose estimation.

Highlights

  • Diverse Solver Set: Supports absolute pose, relative pose, homography, and fundamental matrix estimation from points, lines, and monocular depth estimates.
  • Robust Estimation: Built-in LO-RANSAC and non-linear refinement with multiple loss functions (Huber, Cauchy, etc.).
  • High Performance: Minimal solvers are optimized for speed, with some running in the nanosecond range.
  • COLMAP Compatibility: Uses camera models compatible with COLMAP for seamless integration into 3D reconstruction pipelines.
  • Python Bindings: Full exposure of minimal solvers and robust estimators to Python users.

Related

  • Project
    nianticlabs/map-free-reloc

    A reference implementation for map-free visual relocalization that enables metric pose estimation relative to a single image, removing the need for large 3D maps.

  • Project
    Tau-J/rtmlib

    rtmlib is a lightweight Python library for human and animal pose estimation that wraps RTMPose and ViTPose models. It avoids heavy OpenMMLab dependencies, requiring only numpy, OpenCV and ONNX Runtime (with optional GPU/accelerator back‑ends). The package provides high‑level solution classes (Wholebody, Body, Hand, Animal, Custom, Wholebody3d) that combine a detector (YOLOX, RFDETR, RTMDet) and a pose estimator, plus utilities for drawing skeletons. Models are auto‑downloaded from OpenMMLab or a HuggingFace mirror. Install via pip or from source, and run a Gradio WebUI or use the Python API for single‑image, video, or webcam inference.

  • Project
    jiafeng5513/Evision

    A binocular vision system for camera calibration, disparity mapping, and 3D reconstruction using algorithms like ELAS and ADCensus.

  • Project
    zju3dv/OnePose_Plus_Plus

    OnePose++ is a keypoint-free, one-shot object pose estimation system that determines an object's 3D position and orientation without requiring CAD models.

  • Project
    mprib/caliscope

    A Python toolkit for calibrating multicamera rigs for 3D motion capture, estimating focal length, lens distortion, and relative camera positions.