RecFusion SDK
Reconstruction Class Reference

#include <Reconstruction.h>

Performs colored surface reconstruction from depth data. More...

Detailed Description

Performs colored surface reconstruction from depth data.

Public Member Functions

 Reconstruction (const ReconstructionParams &params)
 Construct surface reconstruction object. More...
 
 ~Reconstruction ()
 Destructor.
 
bool good () const
 Get reconstruction object status. More...
 
bool addFrame (int sensor, const DepthImage &imgDepth, const ColorImage &imgColor, bool *trackingStatus=0, ColorImage *imgScene=0, Mat4 *T=0, Mat4 *sensorT=0, Mat4 *trackedT=0, bool dontForceFirstFrame=false)
 Add frame to reconstruction. More...
 
bool getMesh (Mesh *mesh) const
 Return reconstructed mesh. More...
 

Constructor & Destructor Documentation

◆ Reconstruction()

Reconstruction ( const ReconstructionParams params)

Construct surface reconstruction object.

Parameters
paramsReconstruction parameters.

To check whether the object was created successfully call good() after object creation.

Member Function Documentation

◆ good()

bool good ( ) const

Get reconstruction object status.

Returns
True if reconstruction object is ready for reconstruction, false otherwise.

◆ addFrame()

bool addFrame ( int  sensor,
const DepthImage imgDepth,
const ColorImage imgColor,
bool *  trackingStatus = 0,
ColorImage imgScene = 0,
Mat4 T = 0,
Mat4 sensorT = 0,
Mat4 trackedT = 0,
bool  dontForceFirstFrame = false 
)

Add frame to reconstruction.

Parameters
sensorIndex of sensor from which frame was captured
imgDepthDepth image. Units are assumed to be 0.1 mm.
imgColorColor image
trackingStatusIgnored if zero. True when frame has been tracked successfully, false when track was lost
imgSceneIgnored if zero, otherwise it must point to a color image into which the reconstruction volume will be rendered from the current camera viewpoint. The measurements from the current view are not included in the rendering so on the first call it will return an empty image. Must have same size as depth image.
TIgnored if zero, otherwise it must point to a 4x4 matrix giving the transformation in mm from the camera coordinate system to the center of the reconstruction volume. The camera coordinate system is looking along the positive z-axis with x to the right and y pointing downwards. If T is given the volumePosition in the reconstruction parameters is ignored.
sensorTIgnored if zero, otherwise it must point to a 4x4 matrix specifying the transformation of this sensor relative to the reference sensor.
trackedTIgnored if zero, otherwise it must point to a 4x4 matrix which will return the current pose in mm from the camera coordinate system to the center of the reconstruction volume. This entry is only valid if trackingStatus is true.
Returns
True if frame was processed, false if an error occurred.

◆ getMesh()

bool getMesh ( Mesh mesh) const

Return reconstructed mesh.

The caller has to create the mesh object.

Parameters
meshUser-allocated mesh object
Returns
True if successful, false otherwise

The documentation for this class was generated from the following file:
Search Tab / S to search, Esc to close