Skip to content

GitLab

Commit 37c8cebc authored by Sugon_ldc's avatar Sugon_ldc
Browse files

add new model

parents
Pipeline #318 failed with stages
in 0 seconds
Hide whitespace changes
Inline Side-by-side
Showing with 2291 additions and 0 deletions
configs/body/2d_kpt_sview_rgb_img/associative_embedding/crowdpose/res101_crowdpose_512x512.py 0 → 100755
View file @ 37c8cebc
_base_ = [
'../../../../_base_/default_runtime.py',
'../../../../_base_/datasets/crowdpose.py'
]
checkpoint_config = dict(interval=50)
evaluation = dict(interval=50, metric='mAP', save_best='AP')
optimizer = dict(
type='Adam',
lr=0.0015,
)
optimizer_config = dict(grad_clip=None)
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=0.001,
step=[200, 260])
total_epochs = 300
channel_cfg = dict(
num_output_channels=14,
dataset_joints=14,
dataset_channel=[
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13],
],
inference_channel=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13])
data_cfg = dict(
image_size=512,
base_size=256,
base_sigma=2,
heatmap_size=[128],
num_joints=channel_cfg['dataset_joints'],
dataset_channel=channel_cfg['dataset_channel'],
inference_channel=channel_cfg['inference_channel'],
num_scales=1,
scale_aware_sigma=False,
)
# model settings
model = dict(
type='AssociativeEmbedding',
pretrained='torchvision://resnet101',
backbone=dict(type='ResNet', depth=101),
keypoint_head=dict(
type='AESimpleHead',
in_channels=2048,
num_joints=14,
tag_per_joint=True,
with_ae_loss=[True],
loss_keypoint=dict(
type='MultiLossFactory',
num_joints=14,
num_stages=1,
ae_loss_type='exp',
with_ae_loss=[True],
push_loss_factor=[0.001],
pull_loss_factor=[0.001],
with_heatmaps_loss=[True],
heatmaps_loss_factor=[1.0])),
train_cfg=dict(),
test_cfg=dict(
num_joints=channel_cfg['dataset_joints'],
max_num_people=30,
scale_factor=[1],
with_heatmaps=[True],
with_ae=[True],
project2image=True,
align_corners=False,
nms_kernel=5,
nms_padding=2,
tag_per_joint=True,
detection_threshold=0.1,
tag_threshold=1,
use_detection_val=True,
ignore_too_much=False,
adjust=True,
refine=True,
flip_test=True))
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='BottomUpRandomAffine',
rot_factor=30,
scale_factor=[0.75, 1.5],
scale_type='short',
trans_factor=40),
dict(type='BottomUpRandomFlip', flip_prob=0.5),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(
type='BottomUpGenerateTarget',
sigma=2,
max_num_people=30,
),
dict(
type='Collect',
keys=['img', 'joints', 'targets', 'masks'],
meta_keys=[]),
]
val_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='BottomUpGetImgSize', test_scale_factor=[1]),
dict(
type='BottomUpResizeAlign',
transforms=[
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
]),
dict(
type='Collect',
keys=['img'],
meta_keys=[
'image_file', 'aug_data', 'test_scale_factor', 'base_size',
'center', 'scale', 'flip_index'
]),
]
test_pipeline = val_pipeline
data_root = 'data/crowdpose'
data = dict(
workers_per_gpu=2,
train_dataloader=dict(samples_per_gpu=16),
val_dataloader=dict(samples_per_gpu=1),
test_dataloader=dict(samples_per_gpu=1),
train=dict(
type='BottomUpCrowdPoseDataset',
ann_file=f'{data_root}/annotations/mmpose_crowdpose_trainval.json',
img_prefix=f'{data_root}/images/',
data_cfg=data_cfg,
pipeline=train_pipeline,
dataset_info={{_base_.dataset_info}}),
val=dict(
type='BottomUpCrowdPoseDataset',
ann_file=f'{data_root}/annotations/mmpose_crowdpose_test.json',
img_prefix=f'{data_root}/images/',
data_cfg=data_cfg,
pipeline=val_pipeline,
dataset_info={{_base_.dataset_info}}),
test=dict(
type='BottomUpCrowdPoseDataset',
ann_file=f'{data_root}/annotations/mmpose_crowdpose_test.json',
img_prefix=f'{data_root}/images/',
data_cfg=data_cfg,
pipeline=test_pipeline,
dataset_info={{_base_.dataset_info}}),
)
configs/body/2d_kpt_sview_rgb_img/associative_embedding/crowdpose/res152_crowdpose_512x512.py 0 → 100755
View file @ 37c8cebc
_base_ = [
'../../../../_base_/default_runtime.py',
'../../../../_base_/datasets/crowdpose.py'
]
checkpoint_config = dict(interval=50)
evaluation = dict(interval=50, metric='mAP', save_best='AP')
optimizer = dict(
type='Adam',
lr=0.0015,
)
optimizer_config = dict(grad_clip=None)
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=0.001,
step=[200, 260])
total_epochs = 300
channel_cfg = dict(
num_output_channels=14,
dataset_joints=14,
dataset_channel=[
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13],
],
inference_channel=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13])
data_cfg = dict(
image_size=512,
base_size=256,
base_sigma=2,
heatmap_size=[128],
num_joints=channel_cfg['dataset_joints'],
dataset_channel=channel_cfg['dataset_channel'],
inference_channel=channel_cfg['inference_channel'],
num_scales=1,
scale_aware_sigma=False,
)
# model settings
model = dict(
type='AssociativeEmbedding',
pretrained='torchvision://resnet152',
backbone=dict(type='ResNet', depth=152),
keypoint_head=dict(
type='AESimpleHead',
in_channels=2048,
num_joints=14,
tag_per_joint=True,
with_ae_loss=[True],
loss_keypoint=dict(
type='MultiLossFactory',
num_joints=14,
num_stages=1,
ae_loss_type='exp',
with_ae_loss=[True],
push_loss_factor=[0.001],
pull_loss_factor=[0.001],
with_heatmaps_loss=[True],
heatmaps_loss_factor=[1.0])),
train_cfg=dict(),
test_cfg=dict(
num_joints=channel_cfg['dataset_joints'],
max_num_people=30,
scale_factor=[1],
with_heatmaps=[True],
with_ae=[True],
project2image=True,
align_corners=False,
nms_kernel=5,
nms_padding=2,
tag_per_joint=True,
detection_threshold=0.1,
tag_threshold=1,
use_detection_val=True,
ignore_too_much=False,
adjust=True,
refine=True,
flip_test=True))
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='BottomUpRandomAffine',
rot_factor=30,
scale_factor=[0.75, 1.5],
scale_type='short',
trans_factor=40),
dict(type='BottomUpRandomFlip', flip_prob=0.5),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(
type='BottomUpGenerateTarget',
sigma=2,
max_num_people=30,
),
dict(
type='Collect',
keys=['img', 'joints', 'targets', 'masks'],
meta_keys=[]),
]
val_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='BottomUpGetImgSize', test_scale_factor=[1]),
dict(
type='BottomUpResizeAlign',
transforms=[
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
]),
dict(
type='Collect',
keys=['img'],
meta_keys=[
'image_file', 'aug_data', 'test_scale_factor', 'base_size',
'center', 'scale', 'flip_index'
]),
]
test_pipeline = val_pipeline
data_root = 'data/crowdpose'
data = dict(
workers_per_gpu=2,
train_dataloader=dict(samples_per_gpu=16),
val_dataloader=dict(samples_per_gpu=1),
test_dataloader=dict(samples_per_gpu=1),
train=dict(
type='BottomUpCrowdPoseDataset',
ann_file=f'{data_root}/annotations/mmpose_crowdpose_trainval.json',
img_prefix=f'{data_root}/images/',
data_cfg=data_cfg,
pipeline=train_pipeline,
dataset_info={{_base_.dataset_info}}),
val=dict(
type='BottomUpCrowdPoseDataset',
ann_file=f'{data_root}/annotations/mmpose_crowdpose_test.json',
img_prefix=f'{data_root}/images/',
data_cfg=data_cfg,
pipeline=val_pipeline,
dataset_info={{_base_.dataset_info}}),
test=dict(
type='BottomUpCrowdPoseDataset',
ann_file=f'{data_root}/annotations/mmpose_crowdpose_test.json',
img_prefix=f'{data_root}/images/',
data_cfg=data_cfg,
pipeline=test_pipeline,
dataset_info={{_base_.dataset_info}}),
)
configs/body/2d_kpt_sview_rgb_img/associative_embedding/crowdpose/res50_crowdpose_512x512.py 0 → 100755
View file @ 37c8cebc
_base_ = [
'../../../../_base_/default_runtime.py',
'../../../../_base_/datasets/crowdpose.py'
]
checkpoint_config = dict(interval=50)
evaluation = dict(interval=50, metric='mAP', save_best='AP')
optimizer = dict(
type='Adam',
lr=0.0015,
)
optimizer_config = dict(grad_clip=None)
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=0.001,
step=[200, 260])
total_epochs = 300
channel_cfg = dict(
num_output_channels=14,
dataset_joints=14,
dataset_channel=[
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13],
],
inference_channel=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13])
data_cfg = dict(
image_size=512,
base_size=256,
base_sigma=2,
heatmap_size=[128],
num_joints=channel_cfg['dataset_joints'],
dataset_channel=channel_cfg['dataset_channel'],
inference_channel=channel_cfg['inference_channel'],
num_scales=1,
scale_aware_sigma=False,
)
# model settings
model = dict(
type='AssociativeEmbedding',
pretrained='torchvision://resnet50',
backbone=dict(type='ResNet', depth=50),
keypoint_head=dict(
type='AESimpleHead',
in_channels=2048,
num_joints=14,
tag_per_joint=True,
with_ae_loss=[True],
loss_keypoint=dict(
type='MultiLossFactory',
num_joints=14,
num_stages=1,
ae_loss_type='exp',
with_ae_loss=[True],
push_loss_factor=[0.001],
pull_loss_factor=[0.001],
with_heatmaps_loss=[True],
heatmaps_loss_factor=[1.0],
)),
train_cfg=dict(),
test_cfg=dict(
num_joints=channel_cfg['dataset_joints'],
max_num_people=30,
scale_factor=[1],
with_heatmaps=[True],
with_ae=[True],
project2image=True,
align_corners=False,
nms_kernel=5,
nms_padding=2,
tag_per_joint=True,
detection_threshold=0.1,
tag_threshold=1,
use_detection_val=True,
ignore_too_much=False,
adjust=True,
refine=True,
flip_test=True))
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='BottomUpRandomAffine',
rot_factor=30,
scale_factor=[0.75, 1.5],
scale_type='short',
trans_factor=40),
dict(type='BottomUpRandomFlip', flip_prob=0.5),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(
type='BottomUpGenerateTarget',
sigma=2,
max_num_people=30,
),
dict(
type='Collect',
keys=['img', 'joints', 'targets', 'masks'],
meta_keys=[]),
]
val_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='BottomUpGetImgSize', test_scale_factor=[1]),
dict(
type='BottomUpResizeAlign',
transforms=[
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
]),
dict(
type='Collect',
keys=['img'],
meta_keys=[
'image_file', 'aug_data', 'test_scale_factor', 'base_size',
'center', 'scale', 'flip_index'
]),
]
test_pipeline = val_pipeline
data_root = 'data/crowdpose'
data = dict(
workers_per_gpu=2,
train_dataloader=dict(samples_per_gpu=24),
val_dataloader=dict(samples_per_gpu=1),
test_dataloader=dict(samples_per_gpu=1),
train=dict(
type='BottomUpCrowdPoseDataset',
ann_file=f'{data_root}/annotations/mmpose_crowdpose_trainval.json',
img_prefix=f'{data_root}/images/',
data_cfg=data_cfg,
pipeline=train_pipeline,
dataset_info={{_base_.dataset_info}}),
val=dict(
type='BottomUpCrowdPoseDataset',
ann_file=f'{data_root}/annotations/mmpose_crowdpose_test.json',
img_prefix=f'{data_root}/images/',
data_cfg=data_cfg,
pipeline=val_pipeline,
dataset_info={{_base_.dataset_info}}),
test=dict(
type='BottomUpCrowdPoseDataset',
ann_file=f'{data_root}/annotations/mmpose_crowdpose_test.json',
img_prefix=f'{data_root}/images/',
data_cfg=data_cfg,
pipeline=test_pipeline,
dataset_info={{_base_.dataset_info}}),
)
configs/body/2d_kpt_sview_rgb_img/associative_embedding/mhp/hrnet_mhp.md 0 → 100755
View file @ 37c8cebc
<!-- [ALGORITHM] -->
<details>
<summary align="right"><a href="https://arxiv.org/abs/1611.05424">Associative Embedding (NIPS'2017)</a></summary>
```bibtex
@inproceedings{newell2017associative,
title={Associative embedding: End-to-end learning for joint detection and grouping},
author={Newell, Alejandro and Huang, Zhiao and Deng, Jia},
booktitle={Advances in neural information processing systems},
pages={2277--2287},
year={2017}
}
```
</details>
<!-- [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="https://dl.acm.org/doi/abs/10.1145/3240508.3240509">MHP (ACM MM'2018)</a></summary>
```bibtex
@inproceedings{zhao2018understanding,
title={Understanding humans in crowded scenes: Deep nested adversarial learning and a new benchmark for multi-human parsing},
author={Zhao, Jian and Li, Jianshu and Cheng, Yu and Sim, Terence and Yan, Shuicheng and Feng, Jiashi},
booktitle={Proceedings of the 26th ACM international conference on Multimedia},
pages={792--800},
year={2018}
}
```
</details>
Results on MHP v2.0 validation set without multi-scale test
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [HRNet-w48](/configs/body/2d_kpt_sview_rgb_img/associative_embedding/coco/hrnet_w48_mhp_512x512.py) | 512x512 | 0.583 | 0.895 | 0.666 | 0.656 | 0.931 | [ckpt](https://download.openmmlab.com/mmpose/bottom_up/hrnet_w48_mhp_512x512-85a6ab6f_20201229.pth) | [log](https://download.openmmlab.com/mmpose/bottom_up/hrnet_w48_mhp_512x512_20201229.log.json) |
Results on MHP v2.0 validation set with multi-scale test. 3 default scales (\[2, 1, 0.5\]) are used
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [HRNet-w48](/configs/body/2d_kpt_sview_rgb_img/associative_embedding/coco/hrnet_w48_mhp_512x512.py) | 512x512 | 0.592 | 0.898 | 0.673 | 0.664 | 0.932 | [ckpt](https://download.openmmlab.com/mmpose/bottom_up/hrnet_w48_mhp_512x512-85a6ab6f_20201229.pth) | [log](https://download.openmmlab.com/mmpose/bottom_up/hrnet_w48_mhp_512x512_20201229.log.json) |
configs/body/2d_kpt_sview_rgb_img/associative_embedding/mhp/hrnet_mhp.yml 0 → 100755
View file @ 37c8cebc
Collections:
- Name: HRNet
Paper:
Title: Deep high-resolution representation learning for human pose estimation
URL: http://openaccess.thecvf.com/content_CVPR_2019/html/Sun_Deep_High-Resolution_Representation_Learning_for_Human_Pose_Estimation_CVPR_2019_paper.html
README: https://github.com/open-mmlab/mmpose/blob/master/docs/en/papers/backbones/hrnet.md
Models:
- Config: configs/body/2d_kpt_sview_rgb_img/associative_embedding/coco/hrnet_w48_mhp_512x512.py
In Collection: HRNet
Metadata:
Architecture: &id001
- Associative Embedding
- HRNet
Training Data: MHP
Name: associative_embedding_hrnet_w48_mhp_512x512
Results:
- Dataset: MHP
Metrics:
AP: 0.583
AP@0.5: 0.895
AP@0.75: 0.666
AR: 0.656
AR@0.5: 0.931
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/bottom_up/hrnet_w48_mhp_512x512-85a6ab6f_20201229.pth
- Config: configs/body/2d_kpt_sview_rgb_img/associative_embedding/coco/hrnet_w48_mhp_512x512.py
In Collection: HRNet
Metadata:
Architecture: *id001
Training Data: MHP
Name: associative_embedding_hrnet_w48_mhp_512x512
Results:
- Dataset: MHP
Metrics:
AP: 0.592
AP@0.5: 0.898
AP@0.75: 0.673
AR: 0.664
AR@0.5: 0.932
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/bottom_up/hrnet_w48_mhp_512x512-85a6ab6f_20201229.pth
configs/body/2d_kpt_sview_rgb_img/associative_embedding/mhp/hrnet_w48_mhp_512x512.py 0 → 100755
View file @ 37c8cebc
_base_ = [
'../../../../_base_/default_runtime.py',
'../../../../_base_/datasets/mhp.py'
]
checkpoint_config = dict(interval=50)
evaluation = dict(interval=50, metric='mAP', save_best='AP')
optimizer = dict(
type='Adam',
lr=0.005,
)
optimizer_config = dict(grad_clip=None)
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=0.001,
step=[400, 550])
total_epochs = 600
channel_cfg = dict(
dataset_joints=16,
dataset_channel=[
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15],
],
inference_channel=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15])
data_cfg = dict(
image_size=512,
base_size=256,
base_sigma=2,
heatmap_size=[128],
num_joints=channel_cfg['dataset_joints'],
dataset_channel=channel_cfg['dataset_channel'],
inference_channel=channel_cfg['inference_channel'],
num_scales=1,
scale_aware_sigma=False,
)
# model settings
model = dict(
type='AssociativeEmbedding',
pretrained='https://download.openmmlab.com/mmpose/'
'pretrain_models/hrnet_w48-8ef0771d.pth',
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=(48, 96)),
stage3=dict(
num_modules=4,
num_branches=3,
block='BASIC',
num_blocks=(4, 4, 4),
num_channels=(48, 96, 192)),
stage4=dict(
num_modules=3,
num_branches=4,
block='BASIC',
num_blocks=(4, 4, 4, 4),
num_channels=(48, 96, 192, 384))),
),
keypoint_head=dict(
type='AESimpleHead',
in_channels=48,
num_joints=16,
num_deconv_layers=0,
tag_per_joint=True,
with_ae_loss=[True],
extra=dict(final_conv_kernel=1, ),
loss_keypoint=dict(
type='MultiLossFactory',
num_joints=16,
num_stages=1,
ae_loss_type='exp',
with_ae_loss=[True],
push_loss_factor=[0.01],
pull_loss_factor=[0.01],
with_heatmaps_loss=[True],
heatmaps_loss_factor=[1.0])),
train_cfg=dict(),
test_cfg=dict(
num_joints=channel_cfg['dataset_joints'],
max_num_people=30,
scale_factor=[1],
with_heatmaps=[True],
with_ae=[True],
project2image=True,
align_corners=False,
nms_kernel=5,
nms_padding=2,
tag_per_joint=True,
detection_threshold=0.1,
tag_threshold=1,
use_detection_val=True,
ignore_too_much=False,
adjust=True,
refine=True,
flip_test=True))
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='BottomUpRandomAffine',
rot_factor=30,
scale_factor=[0.75, 1.5],
scale_type='short',
trans_factor=40),
dict(type='BottomUpRandomFlip', flip_prob=0.5),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(
type='BottomUpGenerateTarget',
sigma=2,
max_num_people=30,
),
dict(
type='Collect',
keys=['img', 'joints', 'targets', 'masks'],
meta_keys=[]),
]
val_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='BottomUpGetImgSize', test_scale_factor=[1]),
dict(
type='BottomUpResizeAlign',
transforms=[
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
]),
dict(
type='Collect',
keys=['img'],
meta_keys=[
'image_file', 'aug_data', 'test_scale_factor', 'base_size',
'center', 'scale', 'flip_index'
]),
]
test_pipeline = val_pipeline
data_root = 'data/mhp'
data = dict(
workers_per_gpu=2,
train_dataloader=dict(samples_per_gpu=16),
val_dataloader=dict(samples_per_gpu=1),
test_dataloader=dict(samples_per_gpu=1),
train=dict(
type='BottomUpMhpDataset',
ann_file=f'{data_root}/annotations/mhp_train.json',
img_prefix=f'{data_root}/train/images/',
data_cfg=data_cfg,
pipeline=train_pipeline,
dataset_info={{_base_.dataset_info}}),
val=dict(
type='BottomUpMhpDataset',
ann_file=f'{data_root}/annotations/mhp_val.json',
img_prefix=f'{data_root}/val/images/',
data_cfg=data_cfg,
pipeline=val_pipeline,
dataset_info={{_base_.dataset_info}}),
test=dict(
type='BottomUpMhpDataset',
ann_file=f'{data_root}/annotations/mhp_val.json',
img_prefix=f'{data_root}/val/images/',
data_cfg=data_cfg,
pipeline=test_pipeline,
dataset_info={{_base_.dataset_info}}),
)
configs/body/2d_kpt_sview_rgb_img/cid/coco/hrnet_coco.md 0 → 100755
View file @ 37c8cebc
<!-- [ALGORITHM] -->
<details>
<summary align="right"><a href="https://openaccess.thecvf.com/content/CVPR2022/html/Wang_Contextual_Instance_Decoupling_for_Robust_Multi-Person_Pose_Estimation_CVPR_2022_paper.html">CID (CVPR'2022)</a></summary>
```bibtex
@InProceedings{Wang_2022_CVPR,
author = {Wang, Dongkai and Zhang, Shiliang},
title = {Contextual Instance Decoupling for Robust Multi-Person Pose Estimation},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2022},
pages = {11060-11068}
}
```
</details>
<!-- [DATASET] -->
<details>
<summary align="right"><a href="https://link.springer.com/chapter/10.1007/978-3-319-10602-1_48">COCO (ECCV'2014)</a></summary>
```bibtex
@inproceedings{lin2014microsoft,
title={Microsoft coco: Common objects in context},
author={Lin, Tsung-Yi and Maire, Michael and Belongie, Serge and Hays, James and Perona, Pietro and Ramanan, Deva and Doll{\'a}r, Piotr and Zitnick, C Lawrence},
booktitle={European conference on computer vision},
pages={740--755},
year={2014},
organization={Springer}
}
```
</details>
Results on COCO val2017 without multi-scale test
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [CID](/configs/body/2d_kpt_sview_rgb_img/cid/coco/hrnet_w32_coco_512x512.py) | 512x512 | 0.702 | 0.887 | 0.768 | 0.755 | 0.926 | [ckpt](https://download.openmmlab.com/mmpose/bottom_up/cid/hrnet_w32_coco_512x512-867b9659_20220928.pth) | [log](https://download.openmmlab.com/mmpose/bottom_up/cid/hrnet_w32_coco_512x512_20220928.log.json) |
| [CID](/configs/body/2d_kpt_sview_rgb_img/cid/coco/hrnet_w48_coco_512x512.py) | 512x512 | 0.715 | 0.895 | 0.780 | 0.768 | 0.932 | [ckpt](https://download.openmmlab.com/mmpose/bottom_up/cid/hrnet_w48_coco_512x512-af545767_20221109.pth) | [log](https://download.openmmlab.com/mmpose/bottom_up/cid/hrnet_w48_coco_512x512_20221109.log.json) |
configs/body/2d_kpt_sview_rgb_img/cid/coco/hrnet_coco.yml 0 → 100755
View file @ 37c8cebc
Collections:
- Name: CID
Paper:
Title: Contextual Instance Decoupling for Robust Multi-Person Pose Estimation
URL: https://openaccess.thecvf.com/content/CVPR2022/html/Wang_Contextual_Instance_Decoupling_for_Robust_Multi-Person_Pose_Estimation_CVPR_2022_paper.html
README: https://github.com/open-mmlab/mmpose/blob/master/docs/en/papers/algorithms/cid.md
Models:
- Config: configs/body/2d_kpt_sview_rgb_img/cid/coco/hrnet_w32_coco_512x512.py
In Collection: CID
Metadata:
Architecture: &id001
- CID
Training Data: COCO
Name: cid_hrnet_w32_coco_512x512
Results:
- Dataset: COCO
Metrics:
AP: 0.702
AP@0.5: 0.887
AP@0.75: 0.768
AR: 0.755
AR@0.5: 0.926
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/bottom_up/cid/hrnet_w32_coco_512x512-867b9659_20220928.pth
- Config: configs/body/2d_kpt_sview_rgb_img/cid/coco/hrnet_w48_coco_512x512.py
In Collection: CID
Metadata:
Architecture: *id001
Training Data: COCO
Name: cid_hrnet_w48_coco_512x512
Results:
- Dataset: COCO
Metrics:
AP: 0.715
AP@0.5: 0.895
AP@0.75: 0.78
AR: 0.768
AR@0.5: 0.932
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/bottom_up/cid/hrnet_w48_coco_512x512-af545767_20221109.pth
configs/body/2d_kpt_sview_rgb_img/cid/coco/hrnet_w32_coco_512x512.py 0 → 100755
View file @ 37c8cebc
_base_ = [
'../../../../_base_/default_runtime.py',
'../../../../_base_/datasets/coco.py'
]
checkpoint_config = dict(interval=20)
evaluation = dict(interval=20, metric='mAP', save_best='AP')
optimizer = dict(
type='Adam',
lr=0.001,
)
optimizer_config = dict(grad_clip=None)
# learning policy
lr_config = dict(policy='step', step=[90, 120])
total_epochs = 140
channel_cfg = dict(
dataset_joints=17,
dataset_channel=[
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16],
],
inference_channel=[
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
])
data_cfg = dict(
image_size=512,
base_size=256,
base_sigma=2,
heatmap_size=[128],
num_joints=channel_cfg['dataset_joints'],
dataset_channel=channel_cfg['dataset_channel'],
inference_channel=channel_cfg['inference_channel'],
num_scales=1,
scale_aware_sigma=False,
with_bbox=True,
use_nms=True,
soft_nms=False,
oks_thr=0.8,
)
# model settings
model = dict(
type='CID',
pretrained='https://download.openmmlab.com/mmpose/'
'pretrain_models/hrnet_w32-36af842e.pth',
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),
multiscale_output=True)),
),
keypoint_head=dict(
type='CIDHead',
in_channels=480,
gfd_channels=32,
num_joints=17,
multi_hm_loss_factor=1.0,
single_hm_loss_factor=4.0,
contrastive_loss_factor=1.0,
max_train_instances=200,
prior_prob=0.01),
train_cfg=dict(),
test_cfg=dict(
num_joints=channel_cfg['dataset_joints'],
flip_test=True,
max_num_people=30,
detection_threshold=0.01,
center_pool_kernel=3))
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='BottomUpRandomAffine',
rot_factor=30,
scale_factor=[0.75, 1.5],
scale_type='short',
trans_factor=40),
dict(type='BottomUpRandomFlip', flip_prob=0.5),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(
type='CIDGenerateTarget',
max_num_people=30,
),
dict(
type='Collect',
keys=[
'img', 'multi_heatmap', 'multi_mask', 'instance_coord',
'instance_heatmap', 'instance_mask', 'instance_valid'
],
meta_keys=[]),
]
val_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='BottomUpGetImgSize', test_scale_factor=[1]),
dict(
type='BottomUpResizeAlign',
transforms=[
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
]),
dict(
type='Collect',
keys=['img'],
meta_keys=[
'image_file', 'aug_data', 'test_scale_factor', 'base_size',
'center', 'scale', 'flip_index'
]),
]
test_pipeline = val_pipeline
data_root = 'data/coco'
data = dict(
workers_per_gpu=2,
train_dataloader=dict(samples_per_gpu=20),
val_dataloader=dict(samples_per_gpu=1),
test_dataloader=dict(samples_per_gpu=1),
train=dict(
type='BottomUpCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_train2017.json',
img_prefix=f'{data_root}/train2017/',
data_cfg=data_cfg,
pipeline=train_pipeline,
dataset_info={{_base_.dataset_info}}),
val=dict(
type='BottomUpCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=val_pipeline,
dataset_info={{_base_.dataset_info}}),
test=dict(
type='BottomUpCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=test_pipeline,
dataset_info={{_base_.dataset_info}}),
)
configs/body/2d_kpt_sview_rgb_img/cid/coco/hrnet_w48_coco_512x512.py 0 → 100755
View file @ 37c8cebc
_base_ = [
'../../../../_base_/default_runtime.py',
'../../../../_base_/datasets/coco.py'
]
checkpoint_config = dict(interval=20)
evaluation = dict(interval=20, metric='mAP', save_best='AP')
optimizer = dict(
type='Adam',
lr=0.001,
)
optimizer_config = dict(grad_clip=None)
# learning policy
lr_config = dict(policy='step', step=[90, 120])
total_epochs = 140
channel_cfg = dict(
dataset_joints=17,
dataset_channel=[
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16],
],
inference_channel=[
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
])
data_cfg = dict(
image_size=512,
base_size=256,
base_sigma=2,
heatmap_size=[128],
num_joints=channel_cfg['dataset_joints'],
dataset_channel=channel_cfg['dataset_channel'],
inference_channel=channel_cfg['inference_channel'],
num_scales=1,
scale_aware_sigma=False,
with_bbox=True,
use_nms=True,
soft_nms=False,
oks_thr=0.8,
)
# model settings
model = dict(
type='CID',
pretrained='https://download.openmmlab.com/mmpose/'
'pretrain_models/hrnet_w48-8ef0771d.pth',
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=(48, 96)),
stage3=dict(
num_modules=4,
num_branches=3,
block='BASIC',
num_blocks=(4, 4, 4),
num_channels=(48, 96, 192)),
stage4=dict(
num_modules=3,
num_branches=4,
block='BASIC',
num_blocks=(4, 4, 4, 4),
num_channels=(48, 96, 192, 384),
multiscale_output=True)),
),
keypoint_head=dict(
type='CIDHead',
in_channels=720,
gfd_channels=48,
num_joints=17,
multi_hm_loss_factor=1.0,
single_hm_loss_factor=4.0,
contrastive_loss_factor=1.0,
max_train_instances=200,
prior_prob=0.01),
train_cfg=dict(),
test_cfg=dict(
num_joints=channel_cfg['dataset_joints'],
flip_test=True,
max_num_people=30,
detection_threshold=0.01,
center_pool_kernel=3))
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(
type='BottomUpRandomAffine',
rot_factor=30,
scale_factor=[0.75, 1.5],
scale_type='short',
trans_factor=40),
dict(type='BottomUpRandomFlip', flip_prob=0.5),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(
type='CIDGenerateTarget',
max_num_people=30,
),
dict(
type='Collect',
keys=[
'img', 'multi_heatmap', 'multi_mask', 'instance_coord',
'instance_heatmap', 'instance_mask', 'instance_valid'
],
meta_keys=[]),
]
val_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='BottomUpGetImgSize', test_scale_factor=[1]),
dict(
type='BottomUpResizeAlign',
transforms=[
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
]),
dict(
type='Collect',
keys=['img'],
meta_keys=[
'image_file', 'aug_data', 'test_scale_factor', 'base_size',
'center', 'scale', 'flip_index'
]),
]
test_pipeline = val_pipeline
data_root = 'data/coco'
data = dict(
workers_per_gpu=2,
train_dataloader=dict(samples_per_gpu=20),
val_dataloader=dict(samples_per_gpu=1),
test_dataloader=dict(samples_per_gpu=1),
train=dict(
type='BottomUpCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_train2017.json',
img_prefix=f'{data_root}/train2017/',
data_cfg=data_cfg,
pipeline=train_pipeline,
dataset_info={{_base_.dataset_info}}),
val=dict(
type='BottomUpCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=val_pipeline,
dataset_info={{_base_.dataset_info}}),
test=dict(
type='BottomUpCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=test_pipeline,
dataset_info={{_base_.dataset_info}}),
)
configs/body/2d_kpt_sview_rgb_img/deeppose/README.md 0 → 100755
View file @ 37c8cebc
# DeepPose: Human pose estimation via deep neural networks
## Introduction
<!-- [ALGORITHM] -->
<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_cvpr_2014/html/Toshev_DeepPose_Human_Pose_2014_CVPR_paper.html">DeepPose (CVPR'2014)</a></summary>
```bibtex
@inproceedings{toshev2014deeppose,
title={Deeppose: Human pose estimation via deep neural networks},
author={Toshev, Alexander and Szegedy, Christian},
booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
pages={1653--1660},
year={2014}
}
```
</details>
DeepPose first proposes using deep neural networks (DNNs) to tackle the problem of human pose estimation.
It follows the top-down paradigm, that first detects human bounding boxes and then estimates poses.
It learns to directly regress the human body keypoint coordinates.
configs/body/2d_kpt_sview_rgb_img/deeppose/coco/res101_coco_256x192.py 0 → 100755
View file @ 37c8cebc
_base_ = [
'../../../../_base_/default_runtime.py',
'../../../../_base_/datasets/coco.py'
]
evaluation = dict(interval=10, metric='mAP', save_best='AP')
optimizer = dict(
type='Adam',
lr=5e-4,
)
optimizer_config = dict(grad_clip=None)
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=0.001,
step=[170, 200])
total_epochs = 210
channel_cfg = dict(
num_output_channels=17,
dataset_joints=17,
dataset_channel=[
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16],
],
inference_channel=[
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
])
# model settings
model = dict(
type='TopDown',
pretrained='torchvision://resnet101',
backbone=dict(type='ResNet', depth=101, num_stages=4, out_indices=(3, )),
neck=dict(type='GlobalAveragePooling'),
keypoint_head=dict(
type='DeepposeRegressionHead',
in_channels=2048,
num_joints=channel_cfg['num_output_channels'],
loss_keypoint=dict(type='SmoothL1Loss', use_target_weight=True)),
train_cfg=dict(),
test_cfg=dict(flip_test=True))
data_cfg = dict(
image_size=[192, 256],
heatmap_size=[48, 64],
num_output_channels=channel_cfg['num_output_channels'],
num_joints=channel_cfg['dataset_joints'],
dataset_channel=channel_cfg['dataset_channel'],
inference_channel=channel_cfg['inference_channel'],
soft_nms=False,
nms_thr=1.0,
oks_thr=0.9,
vis_thr=0.2,
use_gt_bbox=False,
det_bbox_thr=0.0,
bbox_file='data/coco/person_detection_results/'
'COCO_val2017_detections_AP_H_56_person.json',
)
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='TopDownGetBboxCenterScale', padding=1.25),
dict(type='TopDownRandomShiftBboxCenter', shift_factor=0.16, prob=0.3),
dict(type='TopDownRandomFlip', flip_prob=0.5),
dict(
type='TopDownHalfBodyTransform',
num_joints_half_body=8,
prob_half_body=0.3),
dict(
type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
dict(type='TopDownAffine'),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(type='TopDownGenerateTargetRegression'),
dict(
type='Collect',
keys=['img', 'target', 'target_weight'],
meta_keys=[
'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
'rotation', 'bbox_score', 'flip_pairs'
]),
]
val_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='TopDownGetBboxCenterScale', padding=1.25),
dict(type='TopDownAffine'),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(
type='Collect',
keys=['img'],
meta_keys=[
'image_file', 'center', 'scale', 'rotation', 'bbox_score',
'flip_pairs'
]),
]
test_pipeline = val_pipeline
data_root = 'data/coco'
data = dict(
samples_per_gpu=64,
workers_per_gpu=2,
val_dataloader=dict(samples_per_gpu=32),
test_dataloader=dict(samples_per_gpu=32),
train=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_train2017.json',
img_prefix=f'{data_root}/train2017/',
data_cfg=data_cfg,
pipeline=train_pipeline,
dataset_info={{_base_.dataset_info}}),
val=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=val_pipeline,
dataset_info={{_base_.dataset_info}}),
test=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=test_pipeline,
dataset_info={{_base_.dataset_info}}),
)
configs/body/2d_kpt_sview_rgb_img/deeppose/coco/res101_coco_256x192_rle.py 0 → 100755
View file @ 37c8cebc
_base_ = [
'../../../../_base_/default_runtime.py',
'../../../../_base_/datasets/coco.py'
]
evaluation = dict(interval=10, metric='mAP', save_best='AP', rle_score=True)
optimizer = dict(
type='Adam',
lr=5e-4,
)
optimizer_config = dict(grad_clip=None)
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=0.001,
step=[170, 200])
total_epochs = 210
channel_cfg = dict(
num_output_channels=17,
dataset_joints=17,
dataset_channel=[
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16],
],
inference_channel=[
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
])
# model settings
model = dict(
type='TopDown',
pretrained='torchvision://resnet101',
backbone=dict(type='ResNet', depth=101, num_stages=4, out_indices=(3, )),
neck=dict(type='GlobalAveragePooling'),
keypoint_head=dict(
type='DeepposeRegressionHead',
in_channels=2048,
num_joints=channel_cfg['num_output_channels'],
loss_keypoint=dict(
type='RLELoss',
use_target_weight=True,
size_average=True,
residual=True),
out_sigma=True),
train_cfg=dict(),
test_cfg=dict(flip_test=True, regression_flip_shift=True))
data_cfg = dict(
image_size=[192, 256],
heatmap_size=[48, 64],
num_output_channels=channel_cfg['num_output_channels'],
num_joints=channel_cfg['dataset_joints'],
dataset_channel=channel_cfg['dataset_channel'],
inference_channel=channel_cfg['inference_channel'],
soft_nms=False,
nms_thr=1.0,
oks_thr=0.9,
vis_thr=0.2,
use_gt_bbox=False,
det_bbox_thr=0.0,
bbox_file='data/coco/person_detection_results/'
'COCO_val2017_detections_AP_H_56_person.json',
)
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='TopDownGetBboxCenterScale', padding=1.25),
dict(type='TopDownRandomShiftBboxCenter', shift_factor=0.16, prob=0.3),
dict(type='TopDownRandomFlip', flip_prob=0.5),
dict(
type='TopDownHalfBodyTransform',
num_joints_half_body=8,
prob_half_body=0.3),
dict(
type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
dict(type='TopDownAffine'),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(type='TopDownGenerateTargetRegression'),
dict(
type='Collect',
keys=['img', 'target', 'target_weight'],
meta_keys=[
'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
'rotation', 'bbox_score', 'flip_pairs'
]),
]
val_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='TopDownGetBboxCenterScale', padding=1.25),
dict(type='TopDownAffine'),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(
type='Collect',
keys=['img'],
meta_keys=[
'image_file', 'center', 'scale', 'rotation', 'bbox_score',
'flip_pairs'
]),
]
test_pipeline = val_pipeline
data_root = 'data/coco'
data = dict(
samples_per_gpu=64,
workers_per_gpu=2,
val_dataloader=dict(samples_per_gpu=32),
test_dataloader=dict(samples_per_gpu=32),
train=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_train2017.json',
img_prefix=f'{data_root}/train2017/',
data_cfg=data_cfg,
pipeline=train_pipeline,
dataset_info={{_base_.dataset_info}}),
val=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=val_pipeline,
dataset_info={{_base_.dataset_info}}),
test=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=test_pipeline,
dataset_info={{_base_.dataset_info}}),
)
configs/body/2d_kpt_sview_rgb_img/deeppose/coco/res152_coco_256x192.py 0 → 100755
View file @ 37c8cebc
_base_ = [
'../../../../_base_/default_runtime.py',
'../../../../_base_/datasets/coco.py'
]
evaluation = dict(interval=10, metric='mAP', save_best='AP')
optimizer = dict(
type='Adam',
lr=5e-4,
)
optimizer_config = dict(grad_clip=None)
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=0.001,
step=[170, 200])
total_epochs = 210
channel_cfg = dict(
num_output_channels=17,
dataset_joints=17,
dataset_channel=[
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16],
],
inference_channel=[
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
])
# model settings
model = dict(
type='TopDown',
pretrained='torchvision://resnet152',
backbone=dict(type='ResNet', depth=152, num_stages=4, out_indices=(3, )),
neck=dict(type='GlobalAveragePooling'),
keypoint_head=dict(
type='DeepposeRegressionHead',
in_channels=2048,
num_joints=channel_cfg['num_output_channels'],
loss_keypoint=dict(type='SmoothL1Loss', use_target_weight=True)),
train_cfg=dict(),
test_cfg=dict(flip_test=True))
data_cfg = dict(
image_size=[192, 256],
heatmap_size=[48, 64],
num_output_channels=channel_cfg['num_output_channels'],
num_joints=channel_cfg['dataset_joints'],
dataset_channel=channel_cfg['dataset_channel'],
inference_channel=channel_cfg['inference_channel'],
soft_nms=False,
nms_thr=1.0,
oks_thr=0.9,
vis_thr=0.2,
use_gt_bbox=False,
det_bbox_thr=0.0,
bbox_file='data/coco/person_detection_results/'
'COCO_val2017_detections_AP_H_56_person.json',
)
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='TopDownGetBboxCenterScale', padding=1.25),
dict(type='TopDownRandomShiftBboxCenter', shift_factor=0.16, prob=0.3),
dict(type='TopDownRandomFlip', flip_prob=0.5),
dict(
type='TopDownHalfBodyTransform',
num_joints_half_body=8,
prob_half_body=0.3),
dict(
type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
dict(type='TopDownAffine'),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(type='TopDownGenerateTargetRegression'),
dict(
type='Collect',
keys=['img', 'target', 'target_weight'],
meta_keys=[
'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
'rotation', 'bbox_score', 'flip_pairs'
]),
]
val_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='TopDownGetBboxCenterScale', padding=1.25),
dict(type='TopDownAffine'),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(
type='Collect',
keys=['img'],
meta_keys=[
'image_file', 'center', 'scale', 'rotation', 'bbox_score',
'flip_pairs'
]),
]
test_pipeline = val_pipeline
data_root = 'data/coco'
data = dict(
samples_per_gpu=64,
workers_per_gpu=2,
val_dataloader=dict(samples_per_gpu=32),
test_dataloader=dict(samples_per_gpu=32),
train=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_train2017.json',
img_prefix=f'{data_root}/train2017/',
data_cfg=data_cfg,
pipeline=train_pipeline,
dataset_info={{_base_.dataset_info}}),
val=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=val_pipeline,
dataset_info={{_base_.dataset_info}}),
test=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=test_pipeline,
dataset_info={{_base_.dataset_info}}),
)
configs/body/2d_kpt_sview_rgb_img/deeppose/coco/res152_coco_256x192_rle.py 0 → 100755
View file @ 37c8cebc
_base_ = [
'../../../../_base_/default_runtime.py',
'../../../../_base_/datasets/coco.py'
]
evaluation = dict(interval=10, metric='mAP', save_best='AP', rle_score=True)
optimizer = dict(
type='Adam',
lr=5e-4,
)
optimizer_config = dict(grad_clip=None)
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=0.001,
step=[170, 200])
total_epochs = 210
channel_cfg = dict(
num_output_channels=17,
dataset_joints=17,
dataset_channel=[
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16],
],
inference_channel=[
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
])
# model settings
model = dict(
type='TopDown',
pretrained='torchvision://resnet152',
backbone=dict(type='ResNet', depth=152, num_stages=4, out_indices=(3, )),
neck=dict(type='GlobalAveragePooling'),
keypoint_head=dict(
type='DeepposeRegressionHead',
in_channels=2048,
num_joints=channel_cfg['num_output_channels'],
loss_keypoint=dict(
type='RLELoss',
use_target_weight=True,
size_average=True,
residual=True),
out_sigma=True),
train_cfg=dict(),
test_cfg=dict(flip_test=True, regression_flip_shift=True))
data_cfg = dict(
image_size=[192, 256],
heatmap_size=[48, 64],
num_output_channels=channel_cfg['num_output_channels'],
num_joints=channel_cfg['dataset_joints'],
dataset_channel=channel_cfg['dataset_channel'],
inference_channel=channel_cfg['inference_channel'],
soft_nms=False,
nms_thr=1.0,
oks_thr=0.9,
vis_thr=0.2,
use_gt_bbox=False,
det_bbox_thr=0.0,
bbox_file='data/coco/person_detection_results/'
'COCO_val2017_detections_AP_H_56_person.json',
)
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='TopDownGetBboxCenterScale', padding=1.25),
dict(type='TopDownRandomShiftBboxCenter', shift_factor=0.16, prob=0.3),
dict(type='TopDownRandomFlip', flip_prob=0.5),
dict(
type='TopDownHalfBodyTransform',
num_joints_half_body=8,
prob_half_body=0.3),
dict(
type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
dict(type='TopDownAffine'),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(type='TopDownGenerateTargetRegression'),
dict(
type='Collect',
keys=['img', 'target', 'target_weight'],
meta_keys=[
'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
'rotation', 'bbox_score', 'flip_pairs'
]),
]
val_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='TopDownGetBboxCenterScale', padding=1.25),
dict(type='TopDownAffine'),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(
type='Collect',
keys=['img'],
meta_keys=[
'image_file', 'center', 'scale', 'rotation', 'bbox_score',
'flip_pairs'
]),
]
test_pipeline = val_pipeline
data_root = 'data/coco'
data = dict(
samples_per_gpu=64,
workers_per_gpu=2,
val_dataloader=dict(samples_per_gpu=32),
test_dataloader=dict(samples_per_gpu=32),
train=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_train2017.json',
img_prefix=f'{data_root}/train2017/',
data_cfg=data_cfg,
pipeline=train_pipeline,
dataset_info={{_base_.dataset_info}}),
val=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=val_pipeline,
dataset_info={{_base_.dataset_info}}),
test=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=test_pipeline,
dataset_info={{_base_.dataset_info}}),
)
configs/body/2d_kpt_sview_rgb_img/deeppose/coco/res152_coco_384x288_rle.py 0 → 100755
View file @ 37c8cebc
_base_ = [
'../../../../_base_/default_runtime.py',
'../../../../_base_/datasets/coco.py'
]
evaluation = dict(interval=10, metric='mAP', save_best='AP', rle_score=True)
optimizer = dict(
type='Adam',
lr=5e-4,
)
optimizer_config = dict(grad_clip=None)
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=0.001,
step=[170, 200])
total_epochs = 210
channel_cfg = dict(
num_output_channels=17,
dataset_joints=17,
dataset_channel=[
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16],
],
inference_channel=[
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
])
# model settings
model = dict(
type='TopDown',
pretrained='torchvision://resnet152',
backbone=dict(type='ResNet', depth=152, num_stages=4, out_indices=(3, )),
neck=dict(type='GlobalAveragePooling'),
keypoint_head=dict(
type='DeepposeRegressionHead',
in_channels=2048,
num_joints=channel_cfg['num_output_channels'],
loss_keypoint=dict(
type='RLELoss',
use_target_weight=True,
size_average=True,
residual=True),
out_sigma=True),
train_cfg=dict(),
test_cfg=dict(flip_test=True, regression_flip_shift=True))
data_cfg = dict(
image_size=[288, 384],
heatmap_size=[72, 96],
num_output_channels=channel_cfg['num_output_channels'],
num_joints=channel_cfg['dataset_joints'],
dataset_channel=channel_cfg['dataset_channel'],
inference_channel=channel_cfg['inference_channel'],
soft_nms=False,
nms_thr=1.0,
oks_thr=0.9,
vis_thr=0.2,
use_gt_bbox=False,
det_bbox_thr=0.0,
bbox_file='data/coco/person_detection_results/'
'COCO_val2017_detections_AP_H_56_person.json',
)
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='TopDownGetBboxCenterScale', padding=1.25),
dict(type='TopDownRandomShiftBboxCenter', shift_factor=0.16, prob=0.3),
dict(type='TopDownRandomFlip', flip_prob=0.5),
dict(
type='TopDownHalfBodyTransform',
num_joints_half_body=8,
prob_half_body=0.3),
dict(
type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
dict(type='TopDownAffine'),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(type='TopDownGenerateTargetRegression'),
dict(
type='Collect',
keys=['img', 'target', 'target_weight'],
meta_keys=[
'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
'rotation', 'bbox_score', 'flip_pairs'
]),
]
val_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='TopDownGetBboxCenterScale', padding=1.25),
dict(type='TopDownAffine'),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(
type='Collect',
keys=['img'],
meta_keys=[
'image_file', 'center', 'scale', 'rotation', 'bbox_score',
'flip_pairs'
]),
]
test_pipeline = val_pipeline
data_root = 'data/coco'
data = dict(
samples_per_gpu=64,
workers_per_gpu=2,
val_dataloader=dict(samples_per_gpu=32),
test_dataloader=dict(samples_per_gpu=32),
train=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_train2017.json',
img_prefix=f'{data_root}/train2017/',
data_cfg=data_cfg,
pipeline=train_pipeline,
dataset_info={{_base_.dataset_info}}),
val=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=val_pipeline,
dataset_info={{_base_.dataset_info}}),
test=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=test_pipeline,
dataset_info={{_base_.dataset_info}}),
)
configs/body/2d_kpt_sview_rgb_img/deeppose/coco/res50_coco_256x192.py 0 → 100755
View file @ 37c8cebc
_base_ = [
'../../../../_base_/default_runtime.py',
'../../../../_base_/datasets/coco.py'
]
evaluation = dict(interval=10, metric='mAP', save_best='AP')
optimizer = dict(
type='Adam',
lr=5e-4,
)
optimizer_config = dict(grad_clip=None)
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=0.001,
step=[170, 200])
total_epochs = 210
channel_cfg = dict(
num_output_channels=17,
dataset_joints=17,
dataset_channel=[
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16],
],
inference_channel=[
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
])
# model settings
model = dict(
type='TopDown',
pretrained='torchvision://resnet50',
backbone=dict(type='ResNet', depth=50, num_stages=4, out_indices=(3, )),
neck=dict(type='GlobalAveragePooling'),
keypoint_head=dict(
type='DeepposeRegressionHead',
in_channels=2048,
num_joints=channel_cfg['num_output_channels'],
loss_keypoint=dict(type='SmoothL1Loss', use_target_weight=True)),
train_cfg=dict(),
test_cfg=dict(flip_test=True))
data_cfg = dict(
image_size=[192, 256],
heatmap_size=[48, 64],
num_output_channels=channel_cfg['num_output_channels'],
num_joints=channel_cfg['dataset_joints'],
dataset_channel=channel_cfg['dataset_channel'],
inference_channel=channel_cfg['inference_channel'],
soft_nms=False,
nms_thr=1.0,
oks_thr=0.9,
vis_thr=0.2,
use_gt_bbox=False,
det_bbox_thr=0.0,
bbox_file='data/coco/person_detection_results/'
'COCO_val2017_detections_AP_H_56_person.json',
)
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='TopDownGetBboxCenterScale', padding=1.25),
dict(type='TopDownRandomShiftBboxCenter', shift_factor=0.16, prob=0.3),
dict(type='TopDownRandomFlip', flip_prob=0.5),
dict(
type='TopDownHalfBodyTransform',
num_joints_half_body=8,
prob_half_body=0.3),
dict(
type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
dict(type='TopDownAffine'),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(type='TopDownGenerateTargetRegression'),
dict(
type='Collect',
keys=['img', 'target', 'target_weight'],
meta_keys=[
'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
'rotation', 'bbox_score', 'flip_pairs'
]),
]
val_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='TopDownGetBboxCenterScale', padding=1.25),
dict(type='TopDownAffine'),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(
type='Collect',
keys=['img'],
meta_keys=[
'image_file', 'center', 'scale', 'rotation', 'bbox_score',
'flip_pairs'
]),
]
test_pipeline = val_pipeline
data_root = 'data/coco'
data = dict(
samples_per_gpu=64,
workers_per_gpu=2,
val_dataloader=dict(samples_per_gpu=32),
test_dataloader=dict(samples_per_gpu=32),
train=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_train2017.json',
img_prefix=f'{data_root}/train2017/',
data_cfg=data_cfg,
pipeline=train_pipeline,
dataset_info={{_base_.dataset_info}}),
val=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=val_pipeline,
dataset_info={{_base_.dataset_info}}),
test=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=test_pipeline,
dataset_info={{_base_.dataset_info}}),
)
configs/body/2d_kpt_sview_rgb_img/deeppose/coco/res50_coco_256x192_rle.py 0 → 100755
View file @ 37c8cebc
_base_ = [
'../../../../_base_/default_runtime.py',
'../../../../_base_/datasets/coco.py'
]
evaluation = dict(interval=10, metric='mAP', save_best='AP', rle_score=True)
optimizer = dict(
type='Adam',
lr=1e-3,
)
optimizer_config = dict(grad_clip=None)
# learning policy
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=0.001,
step=[170, 200])
total_epochs = 210
channel_cfg = dict(
num_output_channels=17,
dataset_joints=17,
dataset_channel=[
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16],
],
inference_channel=[
0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
])
# model settings
model = dict(
type='TopDown',
pretrained='torchvision://resnet50',
backbone=dict(type='ResNet', depth=50, num_stages=4, out_indices=(3, )),
neck=dict(type='GlobalAveragePooling'),
keypoint_head=dict(
type='DeepposeRegressionHead',
in_channels=2048,
num_joints=channel_cfg['num_output_channels'],
loss_keypoint=dict(
type='RLELoss',
use_target_weight=True,
size_average=True,
residual=True),
out_sigma=True),
train_cfg=dict(),
test_cfg=dict(flip_test=True, regression_flip_shift=True))
data_cfg = dict(
image_size=[192, 256],
heatmap_size=[48, 64],
num_output_channels=channel_cfg['num_output_channels'],
num_joints=channel_cfg['dataset_joints'],
dataset_channel=channel_cfg['dataset_channel'],
inference_channel=channel_cfg['inference_channel'],
soft_nms=False,
nms_thr=1.0,
oks_thr=0.9,
vis_thr=0.2,
use_gt_bbox=False,
det_bbox_thr=0.0,
bbox_file='data/coco/person_detection_results/'
'COCO_val2017_detections_AP_H_56_person.json',
)
train_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='TopDownGetBboxCenterScale', padding=1.25),
dict(type='TopDownRandomShiftBboxCenter', shift_factor=0.16, prob=0.3),
dict(type='TopDownRandomFlip', flip_prob=0.5),
dict(
type='TopDownHalfBodyTransform',
num_joints_half_body=8,
prob_half_body=0.3),
dict(
type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
dict(type='TopDownAffine'),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(type='TopDownGenerateTargetRegression'),
dict(
type='Collect',
keys=['img', 'target', 'target_weight'],
meta_keys=[
'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
'rotation', 'bbox_score', 'flip_pairs'
]),
]
val_pipeline = [
dict(type='LoadImageFromFile'),
dict(type='TopDownGetBboxCenterScale', padding=1.25),
dict(type='TopDownAffine'),
dict(type='ToTensor'),
dict(
type='NormalizeTensor',
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]),
dict(
type='Collect',
keys=['img'],
meta_keys=[
'image_file', 'center', 'scale', 'rotation', 'bbox_score',
'flip_pairs'
]),
]
test_pipeline = val_pipeline
data_root = 'data/coco'
data = dict(
samples_per_gpu=64,
workers_per_gpu=2,
val_dataloader=dict(samples_per_gpu=32),
test_dataloader=dict(samples_per_gpu=32),
train=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_train2017.json',
img_prefix=f'{data_root}/train2017/',
data_cfg=data_cfg,
pipeline=train_pipeline,
dataset_info={{_base_.dataset_info}}),
val=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=val_pipeline,
dataset_info={{_base_.dataset_info}}),
test=dict(
type='TopDownCocoDataset',
ann_file=f'{data_root}/annotations/person_keypoints_val2017.json',
img_prefix=f'{data_root}/val2017/',
data_cfg=data_cfg,
pipeline=test_pipeline,
dataset_info={{_base_.dataset_info}}),
)
configs/body/2d_kpt_sview_rgb_img/deeppose/coco/resnet_coco.md 0 → 100755
View file @ 37c8cebc
<!-- [ALGORITHM] -->
<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_cvpr_2014/html/Toshev_DeepPose_Human_Pose_2014_CVPR_paper.html">DeepPose (CVPR'2014)</a></summary>
```bibtex
@inproceedings{toshev2014deeppose,
title={Deeppose: Human pose estimation via deep neural networks},
author={Toshev, Alexander and Szegedy, Christian},
booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
pages={1653--1660},
year={2014}
}
```
</details>
<!-- [BACKBONE] -->
<details>
<summary align="right"><a href="http://openaccess.thecvf.com/content_cvpr_2016/html/He_Deep_Residual_Learning_CVPR_2016_paper.html">ResNet (CVPR'2016)</a></summary>
```bibtex
@inproceedings{he2016deep,
title={Deep residual learning for image recognition},
author={He, Kaiming and Zhang, Xiangyu and Ren, Shaoqing and Sun, Jian},
booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
pages={770--778},
year={2016}
}
```
</details>
<!-- [DATASET] -->
<details>
<summary align="right"><a href="https://link.springer.com/chapter/10.1007/978-3-319-10602-1_48">COCO (ECCV'2014)</a></summary>
```bibtex
@inproceedings{lin2014microsoft,
title={Microsoft coco: Common objects in context},
author={Lin, Tsung-Yi and Maire, Michael and Belongie, Serge and Hays, James and Perona, Pietro and Ramanan, Deva and Doll{\'a}r, Piotr and Zitnick, C Lawrence},
booktitle={European conference on computer vision},
pages={740--755},
year={2014},
organization={Springer}
}
```
</details>
Results on COCO val2017 with detector having human AP of 56.4 on COCO val2017 dataset
| Arch | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
| [deeppose_resnet_50](/configs/body/2d_kpt_sview_rgb_img/deeppose/coco/res50_coco_256x192.py) | 256x192 | 0.526 | 0.816 | 0.586 | 0.638 | 0.887 | [ckpt](https://download.openmmlab.com/mmpose/top_down/deeppose/deeppose_res50_coco_256x192-f6de6c0e_20210205.pth) | [log](https://download.openmmlab.com/mmpose/top_down/deeppose/deeppose_res50_coco_256x192_20210205.log.json) |
| [deeppose_resnet_101](/configs/body/2d_kpt_sview_rgb_img/deeppose/coco/res101_coco_256x192.py) | 256x192 | 0.560 | 0.832 | 0.628 | 0.668 | 0.900 | [ckpt](https://download.openmmlab.com/mmpose/top_down/deeppose/deeppose_res101_coco_256x192-2f247111_20210205.pth) | [log](https://download.openmmlab.com/mmpose/top_down/deeppose/deeppose_res101_coco_256x192_20210205.log.json) |
| [deeppose_resnet_152](/configs/body/2d_kpt_sview_rgb_img/deeppose/coco/res152_coco_256x192.py) | 256x192 | 0.583 | 0.843 | 0.659 | 0.686 | 0.907 | [ckpt](https://download.openmmlab.com/mmpose/top_down/deeppose/deeppose_res152_coco_256x192-7df89a88_20210205.pth) | [log](https://download.openmmlab.com/mmpose/top_down/deeppose/deeppose_res152_coco_256x192_20210205.log.json) |
configs/body/2d_kpt_sview_rgb_img/deeppose/coco/resnet_coco.yml 0 → 100755
View file @ 37c8cebc
Collections:
- Name: ResNet
Paper:
Title: Deep residual learning for image recognition
URL: http://openaccess.thecvf.com/content_cvpr_2016/html/He_Deep_Residual_Learning_CVPR_2016_paper.html
README: https://github.com/open-mmlab/mmpose/blob/master/docs/en/papers/backbones/resnet.md
Models:
- Config: configs/body/2d_kpt_sview_rgb_img/deeppose/coco/res50_coco_256x192.py
In Collection: ResNet
Metadata:
Architecture: &id001
- DeepPose
- ResNet
Training Data: COCO
Name: deeppose_res50_coco_256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.526
AP@0.5: 0.816
AP@0.75: 0.586
AR: 0.638
AR@0.5: 0.887
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/deeppose/deeppose_res50_coco_256x192-f6de6c0e_20210205.pth
- Config: configs/body/2d_kpt_sview_rgb_img/deeppose/coco/res101_coco_256x192.py
In Collection: ResNet
Metadata:
Architecture: *id001
Training Data: COCO
Name: deeppose_res101_coco_256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.56
AP@0.5: 0.832
AP@0.75: 0.628
AR: 0.668
AR@0.5: 0.9
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/deeppose/deeppose_res101_coco_256x192-2f247111_20210205.pth
- Config: configs/body/2d_kpt_sview_rgb_img/deeppose/coco/res152_coco_256x192.py
In Collection: ResNet
Metadata:
Architecture: *id001
Training Data: COCO
Name: deeppose_res152_coco_256x192
Results:
- Dataset: COCO
Metrics:
AP: 0.583
AP@0.5: 0.843
AP@0.75: 0.659
AR: 0.686
AR@0.5: 0.907
Task: Body 2D Keypoint
Weights: https://download.openmmlab.com/mmpose/top_down/deeppose/deeppose_res152_coco_256x192-7df89a88_20210205.pth
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment