Bump versions to v2.22.0

[AronLin]

Changelog

v2.22.0 (24/2/2022)

Highlights

• Support MaskFormer: Per-Pixel Classification is Not All You Need for Semantic Segmentation ([Feature] Add Maskformer to mmdet #7212)
• Support DyHead: Dynamic Head: Unifying Object Detection Heads with Attentions ([Feature] Support DyHead #6823)
• Support Open Images Dataset ([Feature] Support OpenImages Dataset #6331)
• Support TIMM backbone: PyTorch Image Models ([Feature] Support TIMMBackbone #7020)

New Features

• Support MaskFormer ([Feature] Add Maskformer to mmdet #7212)
• Support DyHead ([Feature] Support DyHead #6823)
• Support OpenImages Dataset ([Feature] Support OpenImages Dataset #6331)
• Support TIMM backbone ([Feature] Support TIMMBackbone #7020)
• Support visualization for Panoptic Segmentation ([Feature] Support visualization for Panoptic Segmentation #7041)

Breaking Changes

In order to support the visualization for Panoptic Segmentation, the num_classes can not be None when using the get_palette function to determine whether to use the panoptic palette.

Bug Fixes

• Fix bug for the best checkpoints can not be saved when the key_score is None ([Fix] cannot to save the best checkpoint when the key_score is None #7101)
• Fix MixUp transform filter boxes failing case ([Fix] Fix MixUp transform filter boxes failing case #7080)
• Add missing properties in SABLHead ([Fix] Added missing property in SABLHead #7091)
• Fix bug when NaNs exist in confusion matrix ([Fix] confusion_matrix.py analysis tool NaN errors #7147)
• Fix PALETTE AttributeError in downstream task ([Fix] Fix PALETTE AttributeError in single_gpu_test #7230)

Improvements

• Speed up SimOTA matching ([Enhance] Speed up SimOTA matching. #7098)
• Add Chinese translation of docs_zh-CN/tutorials/init_cfg.md (add Chinese version of init_cfg #7188)

Contributors

A total of 20 developers contributed to this release.
Thanks @ZwwWayne, @hhaAndroid, @RangiLyu, @AronLin, @BIGWangYuDong, @jbwang1997, @zytx121, @chhluo, @shinya7y, @LuooChen, @dvansa, @siatwangmin, @del-zhenwu, @vikashranjan26, @haofanwang, @jamiechoi1995, @HJoonKwon, @yarkable, @zhijian-liu, @RangeKing

[codecov]

Codecov Report

Merging #7240 (f0f09d0) into dev (bb027f1) will decrease coverage by 0.24%.
The diff coverage is 100.00%.

❗ Current head f0f09d0 differs from pull request most recent head 1b425ee. Consider uploading reports for the commit 1b425ee to get more accurate results

@@            Coverage Diff             @@
##              dev    #7240      +/-   ##
==========================================
- Coverage   63.43%   63.19%   -0.25%     
==========================================
  Files         340      340              
  Lines       27469    27279     -190     
  Branches     4640     4640              
==========================================
- Hits        17426    17238     -188     
+ Misses       9115     9114       -1     
+ Partials      928      927       -1     

Flag	Coverage Δ
unittests	63.17% <100.00%> (-0.25%)	⬇️

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

Impacted Files	Coverage Δ
mmdet/__init__.py	100.00% <100.00%> (ø)
mmdet/version.py	61.53% <100.00%> (ø)
mmdet/models/dense_heads/base_mask_head.py	40.00% <0.00%> (-4.45%)	⬇️
mmdet/models/seg_heads/base_semantic_head.py	34.37% <0.00%> (-3.87%)	⬇️
mmdet/models/roi_heads/base_roi_head.py	76.47% <0.00%> (-3.53%)	⬇️
mmdet/core/bbox/coder/yolo_bbox_coder.py	50.00% <0.00%> (-2.95%)	⬇️
mmdet/models/detectors/base.py	50.78% <0.00%> (-2.56%)	⬇️
...panoptic_fusion_heads/base_panoptic_fusion_head.py	52.63% <0.00%> (-2.37%)	⬇️
mmdet/core/bbox/samplers/mask_sampling_result.py	66.66% <0.00%> (-2.09%)	⬇️
mmdet/datasets/pipelines/formatting.py	65.78% <0.00%> (-1.98%)	⬇️
... and 66 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 bb027f1...1b425ee. Read the comment docs.