How to store the positions of codebook entries (visual vocabularies in dictionary) relative to the object center?
Hi I am trying to find all positions of codebook entries relative to the object center and stocked with it for many days. Like the following image:
The steps that I have done till now is extracting keypoints and building the codebook by kmeans clustering (creating visual vocabularies) in python, I do not know to find the center of each visual vocabulary and then calculate the position from the center. Could you please help with this?
Comments
- you cannot do this, from the visual words (BOW) data alone. (there is no position data retained there)
- please show the code, you have, so far