potok/opencv
1
0
Fork 0
mirror of https://github.com/opencv/opencv.git synced 2026-07-21 19:33:03 +04:00
Code Packages Releases Wiki Activity

Add a white background for images which have a transparent background in the calib3d module.

Browse Source 
See:
  - https://github.com/opencv/opencv/pull/28450
  - https://github.com/opencv/opencv/pull/28426
This commit is contained in:
Souriya Trinh
2026-03-05 11:05:07 +01:00
parent fe160f3eed
commit 7f164b5653
6 changed files with 6 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
modules/calib3d/include/opencv2/calib3d.hpp
+6 -6
View File
@@ -220,7 +220,7 @@ Z_w \\
The following figure illustrates the pinhole camera model.
![Pinhole camera model](pics/pinhole_camera_model.png)
![Pinhole camera model](pics/pinhole_camera_model.png) { width=70% }
Real lenses usually have some distortion, mostly radial distortion, and slight tangential distortion.
So, the above model is extended as:
@@ -448,7 +448,7 @@ The inverse of an homogeneous transformation matrix is then:
One can note that the inverse of a 3x3 rotation matrix is directly its matrix transpose.
![Perspective projection, from object to camera frame](pics/pinhole_homogeneous_transformation.png)
![Perspective projection, from object to camera frame](pics/pinhole_homogeneous_transformation.jpg) { width=70% }
This figure summarizes the whole process. The object pose returned for instance by the @ref solvePnP function
or pose from fiducial marker detection is this \f$ {}^{c}\mathbf{T}_o \f$ transformation.
@@ -1012,7 +1012,7 @@ Check @ref tutorial_homography "the corresponding tutorial" for more details
/** @brief Finds an object pose \f$ {}^{c}\mathbf{T}_o \f$ from 3D-2D point correspondences:
![Perspective projection, from object to camera frame](pics/pinhole_homogeneous_transformation.png){ width=50% }
![Perspective projection, from object to camera frame](pics/pinhole_homogeneous_transformation.jpg){ width=50% }
@see @ref calib3d_solvePnP
@@ -1085,7 +1085,7 @@ CV_EXPORTS_W bool solvePnP( InputArray objectPoints, InputArray imagePoints,
/** @brief Finds an object pose \f$ {}^{c}\mathbf{T}_o \f$ from 3D-2D point correspondences using the RANSAC scheme to deal with bad matches.
![Perspective projection, from object to camera frame](pics/pinhole_homogeneous_transformation.png){ width=50% }
![Perspective projection, from object to camera frame](pics/pinhole_homogeneous_transformation.jpg){ width=50% }
@see @ref calib3d_solvePnP
@@ -1147,7 +1147,7 @@ CV_EXPORTS_W bool solvePnPRansac( InputArray objectPoints, InputArray imagePoint
/** @brief Finds an object pose \f$ {}^{c}\mathbf{T}_o \f$ from **3** 3D-2D point correspondences.
![Perspective projection, from object to camera frame](pics/pinhole_homogeneous_transformation.png){ width=50% }
![Perspective projection, from object to camera frame](pics/pinhole_homogeneous_transformation.jpg){ width=50% }
@see @ref calib3d_solvePnP
@@ -1242,7 +1242,7 @@ CV_EXPORTS_W void solvePnPRefineVVS( InputArray objectPoints, InputArray imagePo
/** @brief Finds an object pose \f$ {}^{c}\mathbf{T}_o \f$ from 3D-2D point correspondences.
![Perspective projection, from object to camera frame](pics/pinhole_homogeneous_transformation.png){ width=50% }
![Perspective projection, from object to camera frame](pics/pinhole_homogeneous_transformation.jpg){ width=50% }
@see @ref calib3d_solvePnP