This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
If I understand you correctly, you need to translate between the coordinates of the room and the pixel location of the image. Use findHomography() to understand the transformation and perspectiveTransform() to calculate each new point based on the homography.
| | 2 | provided cite |
If I understand you correctly, you need to translate between the coordinates of the room and the pixel location of the image. Use findHomography() to understand the transformation and perspectiveTransform() to calculate each new point based on the homography.
More at http://answers.opencv.org/question/1858/homography-for-cordinates-of-poins/
| | 3 | added another cite |
If I understand you correctly, you need to translate between the coordinates of the room and the pixel location of the image. Use findHomography() to understand the transformation and perspectiveTransform() to calculate each new point based on the homography.
More at http://answers.opencv.org/question/1858/homography-for-cordinates-of-poins/http://answers.opencv.org/question/1858/homography-for-cordinates-of-poins/ and http://stackoverflow.com/questions/9275567/opencv-homography-transform-a-point-what-is-this-code-doing
| | 4 | added image |
If I understand you correctly, you need to translate between the coordinates of the room (where there is a 0,0 reference somewhere and the room coordinates can be laid out like a grid) and the x,y pixel location of the object in the image. Use findHomography() to understand the transformation and perspectiveTransform() to calculate each new point based on the homography.
This picture helps visualize the problem:
More at http://answers.opencv.org/question/1858/homography-for-cordinates-of-poins/ and http://stackoverflow.com/questions/9275567/opencv-homography-transform-a-point-what-is-this-code-doing
