相机位姿估计和三维重建
Computer Vision Toolbox™ 提供了一整套基础工具，用于从多个视图中估计相机位姿和重建三维场景。这些工具对于运动重建、视觉 SLAM、增强现实和机器人等应用至关重要。您可以使用本质矩阵或基本矩阵来估计相机位姿，还可以基于对极几何计算相机的相对位置。对于已标定的相机，您可以使用透视三点 (P3P) 算法或相机投影矩阵，通过二维到三维点对应关系来确定绝对或相对位姿。有关详细信息，请参阅 estimateEssentialMatrix、estimateFundamentalMatrix 和 estrelpose。
为了优化位姿估计并提高重建精度，该工具箱支持光束法平差技术，可联合或独立地优化相机位姿和三维点位置。有关详细信息，请参阅 bundleAdjustment。
对于三维重建，您可以利用多个视图对点进行三角测量，并管理点轨迹的对应关系。您还可以在图像坐标和世界坐标之间进行转换，这支持投影和反投影任务。有关详细信息，请参阅 pointTrack、triangulate 和 img2world2d。
工具箱提供了使用专用对象来管理和组织图像以及点数据的功能。这些结构能够为大规模重建流程实现位姿图的创建和优化。工具箱还支持在三维空间中可视化相机位置和重建的点云。有关详细信息，请参阅 imageviewset、createPoseGraph 和 plotCamera。
函数
主题
- 多视图运动重建
利用增量式运动重建方法，根据一组二维视图推断场景的三维结构。 (自 R2026a 起)
- Implement Visual SLAM in MATLAB
Understand the visual simultaneous localization and mapping (vSLAM) workflow and how to implement it using MATLAB.
- 坐标系
指定像素索引、空间坐标和三维坐标系。
- 什么是相机标定？
估计图像或视频相机的镜头与图像传感器的参数。
精选示例
Monocular Visual Simultaneous Localization and Mapping
Visual simultaneous localization and mapping (vSLAM).
Stereo Visual Simultaneous Localization and Mapping
Process image data from a stereo camera to build a map of an outdoor environment and estimate the trajectory of the camera.
Monocular Visual-Inertial SLAM
Perform SLAM by combining images captured by a monocular camera with measurements from an IMU sensor.
Visual SLAM with RGB-D Camera
Process RGB-D image data to build a map of an indoor environment and estimate the trajectory of the camera.
Reconstruct Complete 3-D Scene Using Incremental Structure from Motion
Use incremental structure from motion to generate complete 3-D scene by integrating all camera views with the initialized 3-D reconstruction.
Reconstruct 3-D Scene from Geometrically Refined Pair of Initial Views
Initialize 3-D reconstruction by selecting a robust image pair from the geometrically refined view graph.
基于两个视图的运动重建
根据一组二维图像推断场景的三维结构。
Monocular Visual Odometry
Determine location and orientation of a camera by analyzing a sequence of images.
Create View Graph Using Bag of Features
Extract features from images and build a view graph based on visual similarity using bag-of-features approach.
Refine View Graph Using Geometric Verification
Refine the view graph created purely based on visual similarity by verifying geometric consistency between image pairs.
