jcjohnson / jcjohnson/densecap
how to train the model if we don't have ground truth for the bounding box
Open
- Dominant language
- Jupyter Notebook
- Stars
- 1.6k
- Forks
- 424
- PR merge metrics
- No merged PRs in 30d
Description
Hi,
I think that's one of my questions for general object detection method, how to train an end-to-end model if we don't have the ground truth bounding box? in this case, we cannot do box regression or compute region score (?). any ideas?
If there are any existing work on this, please let me know. thanks!
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.