Skip to content

Commit

Permalink
[CodeCamp2023-313]ExLPose data set support (#2778)
Browse files Browse the repository at this point in the history 
[CodeCamp2023-313] ExLPose data set support (#2778)
  • Loading branch information
@Yang-Changhui
Yang-Changhui authored Oct 26, 2023
1 parent 8102383 commit ea6ce0c
Show file tree
Hide file tree
Showing 13 changed files with 1,076 additions and 3 deletions.
125 changes: 125 additions & 0 deletions configs/_base_/datasets/exlpose.py
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,125 @@
dataset_info = dict(
dataset_name='exlpose',
paper_info=dict(
author='Sohyun Lee, Jaesung Rim, Boseung Jeong, Geonu Kim,'
'ByungJu Woo, Haechan Lee, Sunghyun Cho, Suha Kwak',
title='Human Pose Estimation in Extremely Low-Light Conditions',
container='arXiv',
year='2023',
homepage='https://arxiv.org/abs/2303.15410',
),
keypoint_info={
0:
dict(
name='left_shoulder',
id=0,
color=[0, 255, 0],
type='upper',
swap='right_shoulder'),
1:
dict(
name='right_shoulder',
id=1,
color=[255, 128, 0],
type='upper',
swap='left_shoulder'),
2:
dict(
name='left_elbow',
id=2,
color=[0, 255, 0],
type='upper',
swap='right_elbow'),
3:
dict(
name='right_elbow',
id=3,
color=[255, 128, 0],
type='upper',
swap='left_elbow'),
4:
dict(
name='left_wrist',
id=4,
color=[0, 255, 0],
type='upper',
swap='right_wrist'),
5:
dict(
name='right_wrist',
id=5,
color=[255, 128, 0],
type='upper',
swap='left_wrist'),
6:
dict(
name='left_hip',
id=6,
color=[0, 255, 0],
type='lower',
swap='right_hip'),
7:
dict(
name='right_hip',
id=7,
color=[255, 128, 0],
type='lower',
swap='left_hip'),
8:
dict(
name='left_knee',
id=8,
color=[0, 255, 0],
type='lower',
swap='right_knee'),
9:
dict(
name='right_knee',
id=9,
color=[255, 128, 0],
type='lower',
swap='left_knee'),
10:
dict(
name='left_ankle',
id=10,
color=[0, 255, 0],
type='lower',
swap='right_ankle'),
11:
dict(
name='right_ankle',
id=11,
color=[255, 128, 0],
type='lower',
swap='left_ankle'),
12:
dict(name='head', id=12, color=[51, 153, 255], type='upper', swap=''),
13:
dict(name='neck', id=13, color=[51, 153, 255], type='upper', swap='')
},
skeleton_info={
0: dict(link=('head', 'neck'), id=0, color=[51, 153, 255]),
1: dict(link=('neck', 'left_shoulder'), id=1, color=[51, 153, 255]),
2: dict(link=('neck', 'right_shoulder'), id=2, color=[51, 153, 255]),
3: dict(link=('left_shoulder', 'left_elbow'), id=3, color=[0, 255, 0]),
4: dict(link=('left_elbow', 'left_wrist'), id=4, color=[0, 255, 0]),
5: dict(
link=('right_shoulder', 'right_elbow'), id=5, color=[255, 128, 0]),
6:
dict(link=('right_elbow', 'right_wrist'), id=6, color=[255, 128, 0]),
7: dict(link=('neck', 'right_hip'), id=7, color=[51, 153, 255]),
8: dict(link=('neck', 'left_hip'), id=8, color=[51, 153, 255]),
9: dict(link=('right_hip', 'right_knee'), id=9, color=[255, 128, 0]),
10:
dict(link=('right_knee', 'right_ankle'), id=10, color=[255, 128, 0]),
11: dict(link=('left_hip', 'left_knee'), id=11, color=[0, 255, 0]),
12: dict(link=('left_knee', 'left_ankle'), id=12, color=[0, 255, 0]),
},
joint_weights=[
0.2, 0.2, 0.2, 1.3, 1.5, 0.2, 1.3, 1.5, 0.2, 0.2, 0.5, 0.2, 0.2, 0.5
],
sigmas=[
0.079, 0.079, 0.072, 0.072, 0.062, 0.062, 0.107, 0.107, 0.087, 0.087,
0.089, 0.089, 0.079, 0.079
])
38 changes: 38 additions & 0 deletions configs/body_2d_keypoint/topdown_heatmap/exlpose/hrnet_exlpose.md
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,38 @@
<!-- [ALGORITHM] -->

<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_CVPR_2019/html/Sun_Deep_High-Resolution_Representation_Learning_for_Human_Pose_Estimation_CVPR_2019_paper.html">HRNet (CVPR'2019)</a></summary>

```bibtex
@inproceedings{sun2019deep,
title={Deep high-resolution representation learning for human pose estimation},
author={Sun, Ke and Xiao, Bin and Liu, Dong and Wang, Jingdong},
booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
pages={5693--5703},
year={2019}
}
```

</details>

<!-- [DATASET] -->

<details>
<summary align="right"><a href="http://cg.postech.ac.kr/research/ExLPose/">ExLPose (2023)</a></summary>

```bibtex
@inproceedings{ExLPose_2023_CVPR,
title={Human Pose Estimation in Extremely Low-Light Conditions},
author={Sohyun Lee, Jaesung Rim, Boseung Jeong, Geonu Kim, ByungJu Woo, Haechan Lee, Sunghyun Cho, Suha Kwak},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
year={2023}
}
```

</details>

Results on ExLPose-LLA val set with ground-truth bounding boxes

| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [pose_hrnet_w32](/configs/body_2d_keypoint/topdown_heatmap/exlpose/td-hm_hrnet-w32_8xb64-210e_exlpose-256x192.py) | 256x192 | 0.401 | 0.64 | 0.40 | 0.452 | 0.693 | [ckpt](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/exlpose/td-hm_hrnet-w32_8xb64-210e_exlpose-ll-256x192.pth) | [log](https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/exlpose/td-hm_hrnet-w32_8xb64-210e_exlpose-ll-256x192.json) |
18 changes: 18 additions & 0 deletions configs/body_2d_keypoint/topdown_heatmap/exlpose/hrnet_exlpose.yml
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,18 @@
Models:
- Config: configs/body_2d_keypoint/topdown_heatmap/exlpose/td-hm_hrnet-w32_8xb64-210e_exlpose-256x192.py
In Collection: HRNet
Metadata:
Architecture:
- HRNet
Training Data: ExLPose-LL
Name: td-hm_hrnet-w32_8xb64-210e_exlpose-256x192
Results:
- Dataset: ExLPose
Metrics:
AP: 0.401
[email protected]: 0.64
[email protected]: 0.40
AR: 0.452
[email protected]: 0.693
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/v1/body_2d_keypoint/topdown_heatmap/exlpose/td-hm_hrnet-w32_8xb64-210e_exlpose-ll-256x192.pth
149 changes: 149 additions & 0 deletions ...gs/body_2d_keypoint/topdown_heatmap/exlpose/td-hm_hrnet-w32_8xb64-210e_exlpose-256x192.py
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,149 @@
_base_ = ['../../../_base_/default_runtime.py']

# runtime
train_cfg = dict(max_epochs=210, val_interval=10)

# optimizer
optim_wrapper = dict(optimizer=dict(
type='Adam',
lr=5e-4,
))

# learning policy
param_scheduler = [
dict(
type='LinearLR', begin=0, end=500, start_factor=0.001,
by_epoch=False), # warm-up
dict(
type='MultiStepLR',
begin=0,
end=210,
milestones=[170, 200],
gamma=0.1,
by_epoch=True)
]

# automatically scaling LR based on the actual training batch size
auto_scale_lr = dict(base_batch_size=512)

# hooks
default_hooks = dict(checkpoint=dict(save_best='coco/AP', rule='greater'))

# codec settings
codec = dict(
type='MSRAHeatmap', input_size=(192, 256), heatmap_size=(48, 64), sigma=2)

# model settings
model = dict(
type='TopdownPoseEstimator',
data_preprocessor=dict(
type='PoseDataPreprocessor',
mean=[123.675, 116.28, 103.53],
std=[58.395, 57.12, 57.375],
bgr_to_rgb=True),
backbone=dict(
type='HRNet',
in_channels=3,
extra=dict(
stage1=dict(
num_modules=1,
num_branches=1,
block='BOTTLENECK',
num_blocks=(4, ),
num_channels=(64, )),
stage2=dict(
num_modules=1,
num_branches=2,
block='BASIC',
num_blocks=(4, 4),
num_channels=(32, 64)),
stage3=dict(
num_modules=4,
num_branches=3,
block='BASIC',
num_blocks=(4, 4, 4),
num_channels=(32, 64, 128)),
stage4=dict(
num_modules=3,
num_branches=4,
block='BASIC',
num_blocks=(4, 4, 4, 4),
num_channels=(32, 64, 128, 256))),
init_cfg=dict(
type='Pretrained',
checkpoint='https://download.openmmlab.com/mmpose/'
'pretrain_models/hrnet_w32-36af842e.pth'),
),
head=dict(
type='HeatmapHead',
in_channels=32,
out_channels=14,
deconv_out_channels=None,
loss=dict(type='KeypointMSELoss', use_target_weight=True),
decoder=codec),
test_cfg=dict(
flip_test=True,
flip_mode='heatmap',
shift_heatmap=True,
))

# base dataset settings
dataset_type = 'ExlposeDataset'
data_mode = 'topdown'
data_root = 'data/ExLPose/'

# pipelines
train_pipeline = [
dict(type='LoadImage'),
dict(type='GetBBoxCenterScale'),
dict(type='RandomFlip', direction='horizontal'),
dict(type='RandomHalfBody'),
dict(type='RandomBBoxTransform'),
dict(type='TopdownAffine', input_size=codec['input_size']),
dict(type='GenerateTarget', encoder=codec),
dict(type='PackPoseInputs')
]
val_pipeline = [
dict(type='LoadImage'),
dict(type='GetBBoxCenterScale'),
dict(type='TopdownAffine', input_size=codec['input_size']),
dict(type='PackPoseInputs')
]

# data loaders
train_dataloader = dict(
batch_size=64,
num_workers=2,
persistent_workers=True,
sampler=dict(type='DefaultSampler', shuffle=True),
dataset=dict(
type=dataset_type,
data_root=data_root,
data_mode=data_mode,
ann_file='annotations/ExLPose/ExLPose_train_LL.json',
data_prefix=dict(img=''),
pipeline=train_pipeline,
))
val_dataloader = dict(
batch_size=32,
num_workers=2,
persistent_workers=True,
drop_last=False,
sampler=dict(type='DefaultSampler', shuffle=False, round_up=False),
dataset=dict(
type=dataset_type,
data_root=data_root,
data_mode=data_mode,
ann_file='annotations/ExLPose/ExLPose_test_LL-A.json',
data_prefix=dict(img=''),
test_mode=True,
pipeline=val_pipeline,
))
test_dataloader = val_dataloader

# evaluators
val_evaluator = dict(
type='CocoMetric',
ann_file=data_root + 'annotations/ExLPose/ExLPose_test_LL-A.json',
use_area=False)
test_evaluator = val_evaluator
Loading

0 comments on commit ea6ce0c

Please sign in to comment.