Request: Bag of Words from precomputed keypoints and descriptors (Feature #3005)
| Status: | Open | Start date: | 2013-05-03 |
|---|---|---|---|
| Priority: | Low | Due date: | |
| Assignee: | antonio sesto |
% Done: | 0% |
| Category: | features2d | ||
| Target version: | - | ||
| Difficulty: | Pull request: |
Description
Hello,
the API for implementing an algorithm based on the Bag-of-Words (BoW) approach does not contain anything for computing the BoW representation of an image from a pre-computed vocabulary, a pre-computed set of keypoints, and a pre-computed set of descriptors.
Once the vocabulary has been computed, the only available class for producing a BoW representation of an input image is:
[[http://docs.opencv.org/modules/features2d/doc/object_categorization.html?highlight=descriptorextractor#bowimgdescriptorextractor]]
that requires an extractor and a matcher.
Even if it is not hard to overcome this limitation, it would be a nice addition a class where you only need to specify the vocabulary and the descriptors.
History
Updated by Anna Kogan almost 12 years ago
Hello Antonio,
Thank you for reporting the issue. If you could fix the problem on your side, a pull request in our GitHub repo would be highly appreciated!
- Category set to features2d
- Assignee set to antonio sesto
Updated by Maksim Shabunin over 9 years ago
Issue has been transferred to GitHub: https://github.com/Itseez/opencv/issues/4562