Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Feature] Update RTMPose-x wholebody and body models #2498

Merged
merged 1 commit into from
Jun 30, 2023
Merged

[Feature] Update RTMPose-x wholebody and body models #2498

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions projects/rtmpose/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -196,6 +196,7 @@ Feel free to join our community group for more help:
| [RTMPose-l\*](./rtmpose/body_2d_keypoint/rtmpose-l_8xb256-420e_coco-256x192.py) | 256x192 | 76.7 | 95.08 | 70.14 | 27.66 | 4.16 | 18.85 | 3.46 | 45.37 | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-l_simcc-body7_pt-body7_420e-256x192-4dba18fc_20230504.pth) |
| [RTMPose-m\*](./rtmpose/body_2d_keypoint/rtmpose-m_8xb256-420e_coco-384x288.py) | 384x288 | 76.6 | 94.64 | 70.38 | 13.72 | 4.33 | 24.78 | 3.66 | - | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-m_simcc-body7_pt-body7_420e-384x288-65e718c4_20230504.pth) |
| [RTMPose-l\*](./rtmpose/body_2d_keypoint/rtmpose-l_8xb256-420e_coco-384x288.py) | 384x288 | 78.3 | 95.36 | 71.58 | 27.79 | 9.35 | - | 6.05 | - | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-l_simcc-body7_pt-body7_420e-384x288-3f5a1437_20230504.pth) |
| [RTMPose-x\*](./rtmpose/body_2d_keypoint/rtmpose-x_8xb256-700e_coco-384x288.py) | 384x288 | 78.8 | - | - | 49.43 | 17.22 | - | - | - | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-x_simcc-body7_pt-body7_700e-384x288-71d7b7e9_20230629.pth) |

</details>

Expand Down Expand Up @@ -237,6 +238,7 @@ For more details, please refer to [GroupFisher Pruning for RTMPose](./rtmpose/pr
| [RTMPose-m](./rtmpose/wholebody_2d_keypoint/rtmpose-m_8xb64-270e_coco-wholebody-256x192.py) | 256x192 | 58.2 | 67.4 | 2.22 | 13.50 | 4.00 | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-m_simcc-coco-wholebody_pt-aic-coco_270e-256x192-cd5e845c_20230123.pth) |
| [RTMPose-l](./rtmpose/wholebody_2d_keypoint/rtmpose-l_8xb64-270e_coco-wholebody-256x192.py) | 256x192 | 61.1 | 70.0 | 4.52 | 23.41 | 5.67 | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-l_simcc-coco-wholebody_pt-aic-coco_270e-256x192-6f206314_20230124.pth) |
| [RTMPose-l](./rtmpose/wholebody_2d_keypoint/rtmpose-l_8xb32-270e_coco-wholebody-384x288.py) | 384x288 | 64.8 | 73.0 | 10.07 | 44.58 | 7.68 | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-l_simcc-coco-wholebody_pt-aic-coco_270e-384x288-eaeb96c8_20230125.pth) |
| [RTMPose-x](./rtmpose/wholebody_2d_keypoint/rtmpose-x_8xb32-270e_coco-wholebody-384x288.py) | 384x288 | 65.3 | 73.3 | 18.1 | - | - | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-x_simcc-coco-wholebody_pt-body7_270e-384x288-401dfc90_20230629.pth) |

### Animal 2d (17 Keypoints)

Expand Down
2 changes: 2 additions & 0 deletions projects/rtmpose/README_CN.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -187,6 +187,7 @@ RTMPose 是一个长期优化迭代的项目，致力于业务场景下的高性
| [RTMPose-l\*](./rtmpose/body_2d_keypoint/rtmpose-l_8xb256-420e_coco-256x192.py) | 256x192 | 76.7 | 95.08 | 70.14 | 27.66 | 4.16 | 18.85 | 3.46 | 45.37 | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-l_simcc-body7_pt-body7_420e-256x192-4dba18fc_20230504.pth) |
| [RTMPose-m\*](./rtmpose/body_2d_keypoint/rtmpose-m_8xb256-420e_coco-384x288.py) | 384x288 | 76.6 | 94.64 | 70.38 | 13.72 | 4.33 | 24.78 | 3.66 | - | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-m_simcc-body7_pt-body7_420e-384x288-65e718c4_20230504.pth) |
| [RTMPose-l\*](./rtmpose/body_2d_keypoint/rtmpose-l_8xb256-420e_coco-384x288.py) | 384x288 | 78.3 | 95.36 | 71.58 | 27.79 | 9.35 | - | 6.05 | - | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-l_simcc-body7_pt-body7_420e-384x288-3f5a1437_20230504.pth) |
| [RTMPose-x\*](./rtmpose/body_2d_keypoint/rtmpose-x_8xb256-700e_coco-384x288.py) | 384x288 | 78.8 | - | - | 49.43 | 17.22 | - | - | - | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-x_simcc-body7_pt-body7_700e-384x288-71d7b7e9_20230629.pth) |

</details>

Expand Down Expand Up @@ -228,6 +229,7 @@ RTMPose 是一个长期优化迭代的项目，致力于业务场景下的高性
| [RTMPose-m](./rtmpose/wholebody_2d_keypoint/rtmpose-m_8xb64-270e_coco-wholebody-256x192.py) | 256x192 | 58.2 | 67.4 | 2.22 | 13.50 | 4.00 | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-m_simcc-coco-wholebody_pt-aic-coco_270e-256x192-cd5e845c_20230123.pth) |
| [RTMPose-l](./rtmpose/wholebody_2d_keypoint/rtmpose-l_8xb64-270e_coco-wholebody-256x192.py) | 256x192 | 61.1 | 70.0 | 4.52 | 23.41 | 5.67 | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-l_simcc-coco-wholebody_pt-aic-coco_270e-256x192-6f206314_20230124.pth) |
| [RTMPose-l](./rtmpose/wholebody_2d_keypoint/rtmpose-l_8xb32-270e_coco-wholebody-384x288.py) | 384x288 | 64.8 | 73.0 | 10.07 | 44.58 | 7.68 | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-l_simcc-coco-wholebody_pt-aic-coco_270e-384x288-eaeb96c8_20230125.pth) |
| [RTMPose-x](./rtmpose/wholebody_2d_keypoint/rtmpose-x_8xb32-270e_coco-wholebody-384x288.py) | 384x288 | 65.3 | 73.3 | 18.1 | - | - | [Model](https://download.openmmlab.com/mmpose/v1/projects/rtmposev1/rtmpose-x_simcc-coco-wholebody_pt-body7_270e-384x288-401dfc90_20230629.pth) |

### 动物 2d 关键点 (17 Keypoints)

Expand Down
238 changes: 238 additions & 0 deletions projects/rtmpose/rtmpose/body_2d_keypoint/rtmpose-x_8xb256-700e_coco-384x288.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,238 @@
_base_ = ['mmpose::_base_/default_runtime.py']

# common setting
num_keypoints = 17
input_size = (288, 384)

# runtime
max_epochs = 700
stage2_num_epochs = 20
base_lr = 4e-3
train_batch_size = 256
val_batch_size = 64

train_cfg = dict(max_epochs=max_epochs, val_interval=10)
randomness = dict(seed=21)

# optimizer
optim_wrapper = dict(
type='OptimWrapper',
optimizer=dict(type='AdamW', lr=base_lr, weight_decay=0.05),
clip_grad=dict(max_norm=35, norm_type=2),
paramwise_cfg=dict(
norm_decay_mult=0, bias_decay_mult=0, bypass_duplicate=True))

# learning rate
param_scheduler = [
dict(
type='LinearLR',
start_factor=1.0e-5,
by_epoch=False,
begin=0,
end=1000),
dict(
type='CosineAnnealingLR',
eta_min=base_lr * 0.05,
begin=max_epochs // 2,
end=max_epochs,
T_max=max_epochs // 2,
by_epoch=True,
convert_to_iter_based=True),
]

# automatically scaling LR based on the actual training batch size
auto_scale_lr = dict(base_batch_size=1024)

# codec settings
codec = dict(
type='SimCCLabel',
input_size=input_size,
sigma=(6., 6.93),
simcc_split_ratio=2.0,
normalize=False,
use_dark=False)

# 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(
_scope_='mmdet',
type='CSPNeXt',
arch='P5',
expand_ratio=0.5,
deepen_factor=1.33,
widen_factor=1.28,
out_indices=(4, ),
channel_attention=True,
norm_cfg=dict(type='SyncBN'),
act_cfg=dict(type='SiLU'),
init_cfg=dict(
type='Pretrained',
prefix='backbone.',
checkpoint='https://download.openmmlab.com/mmpose/v1/projects/'
'rtmposev1/cspnext-x_udp-body7_210e-384x288-d28b58e6_20230529.pth' # noqa
)),
head=dict(
type='RTMCCHead',
in_channels=1280,
out_channels=num_keypoints,
input_size=codec['input_size'],
in_featuremap_size=tuple([s // 32 for s in codec['input_size']]),
simcc_split_ratio=codec['simcc_split_ratio'],
final_layer_kernel_size=7,
gau_cfg=dict(
hidden_dims=256,
s=128,
expansion_factor=2,
dropout_rate=0.,
drop_path=0.,
act_fn='SiLU',
use_rel_bias=False,
pos_enc=False),
loss=dict(
type='KLDiscretLoss',
use_target_weight=True,
beta=10.,
label_softmax=True),
decoder=codec),
test_cfg=dict(flip_test=True))

# base dataset settings
dataset_type = 'CocoDataset'
data_mode = 'topdown'
data_root = 'data/coco/'

backend_args = dict(backend='local')
# backend_args = dict(
# backend='petrel',
# path_mapping=dict({
# f'{data_root}': 's3://openmmlab/datasets/detection/coco/',
# f'{data_root}': 's3://openmmlab/datasets/detection/coco/'
# }))

# pipelines
train_pipeline = [
dict(type='LoadImage', backend_args=backend_args),
dict(type='GetBBoxCenterScale'),
dict(type='RandomFlip', direction='horizontal'),
dict(type='RandomHalfBody'),
dict(
type='RandomBBoxTransform', scale_factor=[0.5, 1.5], rotate_factor=90),
dict(type='TopdownAffine', input_size=codec['input_size']),
dict(type='PhotometricDistortion'),
dict(
type='Albumentation',
transforms=[
dict(type='Blur', p=0.1),
dict(type='MedianBlur', p=0.1),
dict(
type='CoarseDropout',
max_holes=1,
max_height=0.4,
max_width=0.4,
min_holes=1,
min_height=0.2,
min_width=0.2,
p=1.),
]),
dict(type='GenerateTarget', encoder=codec),
dict(type='PackPoseInputs')
]
val_pipeline = [
dict(type='LoadImage', backend_args=backend_args),
dict(type='GetBBoxCenterScale'),
dict(type='TopdownAffine', input_size=codec['input_size']),
dict(type='PackPoseInputs')
]

train_pipeline_stage2 = [
dict(type='LoadImage', backend_args=backend_args),
dict(type='GetBBoxCenterScale'),
dict(type='RandomFlip', direction='horizontal'),
dict(type='RandomHalfBody'),
dict(
type='RandomBBoxTransform',
shift_factor=0.,
scale_factor=[0.5, 1.5],
rotate_factor=90),
dict(type='TopdownAffine', input_size=codec['input_size']),
dict(type='mmdet.YOLOXHSVRandomAug'),
dict(
type='Albumentation',
transforms=[
dict(type='Blur', p=0.1),
dict(type='MedianBlur', p=0.1),
dict(
type='CoarseDropout',
max_holes=1,
max_height=0.4,
max_width=0.4,
min_holes=1,
min_height=0.2,
min_width=0.2,
p=0.5),
]),
dict(type='GenerateTarget', encoder=codec),
dict(type='PackPoseInputs')
]

# data loaders
train_dataloader = dict(
batch_size=train_batch_size,
num_workers=10,
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/person_keypoints_train2017.json',
data_prefix=dict(img='train2017/'),
pipeline=train_pipeline,
))
val_dataloader = dict(
batch_size=val_batch_size,
num_workers=10,
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/person_keypoints_val2017.json',
bbox_file=f'{data_root}person_detection_results/'
'COCO_val2017_detections_AP_H_56_person.json',
data_prefix=dict(img='val2017/'),
test_mode=True,
pipeline=val_pipeline,
))
test_dataloader = val_dataloader

# hooks
default_hooks = dict(
checkpoint=dict(save_best='coco/AP', rule='greater', max_keep_ckpts=1))

custom_hooks = [
dict(
type='EMAHook',
ema_type='ExpMomentumEMA',
momentum=0.0002,
update_buffers=True,
priority=49),
dict(
type='mmdet.PipelineSwitchHook',
switch_epoch=max_epochs - stage2_num_epochs,
switch_pipeline=train_pipeline_stage2)
]

# evaluators
val_evaluator = dict(
type='CocoMetric',
ann_file=data_root + 'annotations/person_keypoints_val2017.json')
test_evaluator = val_evaluator
Loading