-
Notifications
You must be signed in to change notification settings - Fork 7.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add ignore regions to reduce effect from synonymous objects #1909
Comments
It's unlikely we'll add to existing models large features that are specifically tailored to properties of certain datasets. Such features are expected to be implemented separately as new modules in users' code, or potentially in If there are any good utilities (e.g., IOU vs IOA) or more general improvements that can simplify the above process or are part of the abovementioned feature, they can be discussed separately as individual feature requests. |
I just noticed #1910 and it gave a better idea of what's needed to implement this. For reasons above (and more in docs) we may not be able to accept it in its current form, but some independent pieces of it can be polished and included. For example, if we for some reason need to implement this model, we'll take the following approach:
Whether (3) can go into the existing models, or should be implemented separately outside detectron2 will depend on its complexity, which also relies on work done in (1) and (2). |
Thanks for your replying and suggestions, I will try to split those features and re-implement them. |
I have re-implemented those three features, and pull-request the first two features. |
Summary: ref to issues in #1909 and pr in #1910 The matcher ignore area strategy split into 3 parts: 1. pump the extra data into the model 2. **utility functions IoA** 3. ignore strategy for RPN here is the second implementation, I extract intersection calculation from pairwise_iou function, and implement pairwise_ioa based on this, also add relevant testcase for IoA. Pull Request resolved: #1921 Reviewed By: rbgirshick Differential Revision: D23286904 Pulled By: ppwwyyxx fbshipit-source-id: 8bea029263adcb22e2fffebcf90d0c15748e5913
We can proceed as follows: Let's use an extra
|
It's clear, I will try to implement them after the holidays |
some datasets like CrowdHuman contains 'ignore regions' bounding boxes annotation, there are 100k ignore bounding boxes in CrowdHuman, when we train a detector based on this dataset, we may want to reduce the effect from human doll, human billboard or dense crowd human. these part of regions should not consider as positive and negative supervise, bug I notice Detectron2 don't have a feature to process 'ignore' annotation during dataloader and matcher, this feature cound be used for lots of projects which need ignore labels
The text was updated successfully, but these errors were encountered: