Structure from Motion

3-D reconstruction from multiple views

Structure from Motion (SfM) is the process of estimating the 3-D structure of a scene from a set of 2-D images. For more details, see Implement Visual SLAM in MATLAB.

Apps

Camera Calibrator	Estimate geometric parameters of a single camera
Stereo Camera Calibrator	Estimate geometric parameters of a stereo camera

Functions

expand all

Detect Features

`detectBRISKFeatures`	Detect BRISK features
`detectFASTFeatures`	Detect corners using FAST algorithm
`detectHarrisFeatures`	Detect corners using Harris–Stephens algorithm
`detectMinEigenFeatures`	Detect corners using minimum eigenvalue algorithm
`detectMSERFeatures`	Detect MSER features
`detectSIFTFeatures`	Detect scale invariant feature transform (SIFT) features
`detectSURFFeatures`	Detect SURF features

Match Features

`extractFeatures`	Extract interest point descriptors
`matchFeatures`	Find matching features
`matchFeaturesInRadius`	Find matching features within specified radius
`vision.PointTracker`	Track points in video using Kanade-Lucas-Tomasi (KLT) algorithm

Estimate 3-D Structure

Store Image and Camera Data

`imageviewset`	Manage data for structure-from-motion, visual odometry, and visual SLAM
`worldpointset`	Manage 3-D to 2-D point correspondences
`cameraIntrinsics`	Object for storing intrinsic camera parameters
`rigidtform3d`	3-D rigid geometric transformation
`affinetform3d`	3-D affine geometric transformation

Estimate Camera Poses

`estimateEssentialMatrix`	Estimate essential matrix from corresponding points in a pair of images
`estimateFundamentalMatrix`	Estimate fundamental matrix from corresponding points in stereo images
`estworldpose`	Estimate camera pose from 3-D to 2-D point correspondences
`estrelpose`	Calculate relative rotation and translation between camera poses

Triangulate Image Points

`pointTrack`	Object for storing matching points from multiple views
`findTracks`	Find matched points across multiple views
`triangulate`	3-D locations of undistorted matching points in stereo images
`triangulateMultiview`	3-D locations of world points matched across multiple images

Optimize Camera Poses and 3-D Points

`bundleAdjustment`	Adjust collection of 3-D points and camera poses
`bundleAdjustmentMotion`	Adjust collection of 3-D points and camera poses using motion-only bundle adjustment
`bundleAdjustmentStructure`	Refine 3-D points using structure-only bundle adjustment

Visualize Results

`stereoAnaglyph`	Create red-cyan anaglyph from stereo pair of images
`pcshow`	Plot 3-D point cloud
`plotCamera`	Plot a camera in 3-D coordinates
`showMatchedFeatures`	Display corresponding feature points

Conversions

`rotmat2vec3d`	Convert 3-D rotation matrix to rotation vector
`rotvec2mat3d`	Convert 3-D rotation vector to rotation matrix

Topics

Apps for Camera Calibration

Using the Single Camera Calibrator App
Estimate camera intrinsics, extrinsics, and lens distortion parameters.
Using the Stereo Camera Calibrator App
Calibrate a stereo camera, which you can then use to recover depth from images.

Visual Odometry

Monocular Visual Odometry
Determine location and orientation of a camera by analyzing a sequence of images.
Monocular Visual Simultaneous Localization and Mapping
Visual simultaneous localization and mapping (vSLAM).

Basics

Coordinate Systems
Specify pixel Indices, spatial coordinates, and 3-D coordinate systems
Point Feature Types
Choose functions that return and accept points objects for several types of features.
Local Feature Detection and Extraction
Learn the benefits and applications of local feature detection and extraction.
Structure from Motion Overview
Estimate three-dimensional structures from two-dimensional image sequences
Implement Visual SLAM in MATLAB
Understand the visual simultaneous localization and mapping (vSLAM) workflow and how to implement it using MATLAB.

Featured Examples

Monocular Visual Simultaneous Localization and Mapping

Visual simultaneous localization and mapping (vSLAM).

Open Live Script

Structure From Motion From Multiple Views

Reconstruct a 3-D scene from a sequence of 2-D views taken with a camera calibrated.

Open Live Script

Structure From Motion From Two Views

Estimate 3-D structure of a scene from a set of 2-D images.

Open Live Script