[Feature] Support GLIP

[okotaku]

Motivation

code: https://github.com/microsoft/GLIP
paper: https://arxiv.org/abs/2112.03857

Checklist

Before PR:

• Pre-commit or other linting tools are used to fix the potential lint issues.
• Bug fixes are fully covered by unit tests, the case that causes the bug should be added in the unit tests.
• The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness.
• The documentation has been modified accordingly, like docstring or example tutorials.

After PR:

• If the modification has potential influence on downstream or other related projects, this PR should be tested with those projects, like MMDet or MMSeg.
• CLA has been signed and all committers have signed the CLA in this PR.

[codecov]

Codecov Report

Patch coverage has no change and project coverage change: +0.80 🎉

Comparison is base (c9a0cb0) 84.37% compared to head (67edd21) 85.17%.

❗ Current head 67edd21 differs from pull request most recent head fb768d5. Consider uploading reports for the commit fb768d5 to get more accurate results

Additional details and impacted files

@@            Coverage Diff             @@
##              dev    #1308      +/-   ##
==========================================
+ Coverage   84.37%   85.17%   +0.80%     
==========================================
  Files         142      230      +88     
  Lines        9925    17492    +7567     
  Branches     1621     2742    +1121     
==========================================
+ Hits         8374    14899    +6525     
- Misses       1277     2079     +802     
- Partials      274      514     +240     

Flag	Coverage Δ
unittests	85.17% <ø> (+0.80%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

see 372 files with indirect coverage changes

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[yingfhu]

Are you intend to use the GLIP swin backbone as the pretrained model in classification? It seems like the original paper does not attempt such classification task, but more for detection. Have you done some experiments to prove its performance, otherwise it seems unnecessary to support extra pretrains of swin

[Ezra-Yu]

LGTM.