[Fix] Include images without instance into bottom-up datasets for validation

[Ben-Louis]

Motivation

The initial implementation of bottom-up datasets fails to take into account images lacking instance annotations, thus deviating from the standard evaluation protocol and resulting in an inflated accuracy score.

Modification

1. Modify _load_annotations in BaseCocoStyleDataset, MpiiDataset, MpiiTrbDataset and CocoWholeBodyDataset to support returning both instance list and image list
2. Modify BaseCocoStyleDataset._get_bottomup_data_infos to add images without instance into data_list_bu
3. Update CocoMetric to adapt the above modifications
4. Update unittests

BC-breaking (Optional)

Use cases (Optional)

Checklist

Before PR:

• I have read and followed the workflow indicated in the CONTRIBUTING.md to create this PR.
• Pre-commit or linting tools indicated in CONTRIBUTING.md are used to fix the potential lint issues.
• Bug fixes are covered by unit tests, the case that causes the bug should be added in the unit tests.
• New functionalities are covered by complete unit tests. If not, please add more unit tests to ensure correctness.
• The documentation has been modified accordingly, including docstring or example tutorials.

After PR:

• CLA has been signed and all committers have signed the CLA in this PR.

[codecov]

Codecov Report

Base: 80.22% // Head: 80.28% // Increases project coverage by +0.05% 🎉

Coverage data is based on head (b09419b) compared to base (cec0a04).
Patch coverage: 90.00% of modified lines in pull request are covered.

❗ Current head b09419b differs from pull request most recent head ac1f68f. Consider uploading reports for the commit ac1f68f to get more accurate results

Additional details and impacted files

@@             Coverage Diff             @@
##           dev-1.x    #1936      +/-   ##
===========================================
+ Coverage    80.22%   80.28%   +0.05%     
===========================================
  Files          209      209              
  Lines        12417    12464      +47     
  Branches      2100     2105       +5     
===========================================
+ Hits          9962    10007      +45     
- Misses        2004     2005       +1     
- Partials       451      452       +1     

Flag	Coverage Δ
unittests	80.28% <90.00%> (+0.05%)	⬆️

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

Impacted Files	Coverage Δ
mmpose/evaluation/metrics/coco_metric.py	95.95% <25.00%> (-0.98%)	⬇️
mmpose/datasets/datasets/body/mpii_dataset.py	65.57% <83.33%> (+1.28%)	⬆️
mmpose/datasets/datasets/body/mpii_trb_dataset.py	92.10% <84.61%> (-1.84%)	⬇️
mmpose/datasets/__init__.py	100.00% <100.00%> (ø)
mmpose/datasets/dataset_wrappers.py	93.61% <100.00%> (+8.16%)	⬆️
.../datasets/datasets/base/base_coco_style_dataset.py	89.28% <100.00%> (+0.75%)	⬆️
...asets/datasets/hand/coco_wholebody_hand_dataset.py	100.00% <100.00%> (ø)
mmpose/datasets/transforms/common_transforms.py	85.21% <0.00%> (-0.71%)	⬇️
mmpose/apis/webcam/nodes/base_visualizer_node.py	100.00% <0.00%> (ø)
mmpose/apis/webcam/utils/buffer.py	91.66% <0.00%> (+0.14%)	⬆️
... and 11 more

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 at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.