[Algorithm] Support Circular Smooth Label (CSL, ECCV'20)

[yangxue0827]

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.

Motivation

Support Circular Smooth Label (CSL, ECCV'20) for #117.
Special thanks to liuyanyi for his help.

Modification

• Add new head: csl_rotated_retina_head
• Add new loss: smooth_focal_loss
• Add new coder: angle_coder
• Add new configs: rotated_retinanet_obb_csl_gaussian_r50_fpn_fp16_1x_dota_le90
• Update documentations: model_zoo_en, model_zoo_zh-cn, README, README_zh-CN

BC-breaking (Optional)

Does the modification introduce changes that break the back-compatibility of the downstream repos?
If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR.

Use cases (Optional)

If this PR introduces a new feature, it is better to list some use cases here, and update the documentation.

Checklist

1. Pre-commit or other linting tools are used to fix the potential lint issues.
2. The modification is covered by complete unit tests. If not, please add more unit test to ensure the correctness.
3. The documentation has been modified accordingly, like docstring or example tutorials.

[codecov-commenter]

Codecov Report

Merging #153 (7349f30) into dev (4e39a45) will decrease coverage by 0.37%.
The diff coverage is 16.54%.

@@            Coverage Diff             @@
##              dev     #153      +/-   ##
==========================================
- Coverage   28.67%   28.30%   -0.38%     
==========================================
  Files         102      108       +6     
  Lines        6604     6967     +363     
  Branches      990     1048      +58     
==========================================
+ Hits         1894     1972      +78     
- Misses       4646     4927     +281     
- Partials       64       68       +4     

Flag	Coverage Δ
unittests	28.29% <16.54%> (-0.38%)	⬇️

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

Impacted Files	Coverage Δ
mmrotate/apis/inference.py	0.00% <ø> (ø)
mmrotate/core/anchor/anchor_generator.py	75.00% <ø> (ø)
mmrotate/datasets/pipelines/loading.py	30.43% <ø> (ø)
...ate/models/dense_heads/kfiou_rotate_retina_head.py	30.00% <ø> (ø)
...els/dense_heads/kfiou_rotate_retina_refine_head.py	25.00% <ø> (ø)
mmrotate/models/dense_heads/oriented_rpn_head.py	12.17% <ø> (ø)
mmrotate/models/dense_heads/rotated_anchor_head.py	10.12% <0.00%> (-0.64%)	⬇️
...e/models/dense_heads/rotated_retina_refine_head.py	25.00% <ø> (ø)
mmrotate/models/dense_heads/rotated_rpn_head.py	12.29% <ø> (ø)
mmrotate/models/dense_heads/sam_reppoints_head.py	8.65% <0.00%> (-0.03%)	⬇️
... and 28 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 7c9a592...7349f30. Read the comment docs.

[zytx121]

LGTM

[jbwang1997]

Paper URL and README are incorrect. And need a Code field.

[zytx121]

@jbwang1997 We previously referenced MMPose's metafile.yml setting, so there is no code field. Maybe, we could propose another PR later to modify it uniformly.

[jbwang1997]

Have a little metafile.yml error. Others look good to me.

[zytx121]

The model's weights and logs have been uploaded.