add new model

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split2.py

+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+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
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet152',
+    backbone=dict(type='ResNet', depth=152),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 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'])
+
+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='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='TopDownGenerateTarget', sigma=2),
+    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/horse10'
+data = dict(
+    samples_per_gpu=32,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split3.py

+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+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
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet152',
+    backbone=dict(type='ResNet', depth=152),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 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'])
+
+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='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='TopDownGenerateTarget', sigma=2),
+    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/horse10'
+data = dict(
+    samples_per_gpu=32,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split1.py

+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+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
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet50',
+    backbone=dict(type='ResNet', depth=50),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 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'])
+
+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='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='TopDownGenerateTarget', sigma=2),
+    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/horse10'
+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='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split2.py

+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+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
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet50',
+    backbone=dict(type='ResNet', depth=50),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 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'])
+
+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='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='TopDownGenerateTarget', sigma=2),
+    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/horse10'
+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='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split3.py

+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+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
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet50',
+    backbone=dict(type='ResNet', depth=50),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 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'])
+
+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='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='TopDownGenerateTarget', sigma=2),
+    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/horse10'
+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='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/resnet_horse10.md

+<!-- [ALGORITHM] -->
+
+<details>
+<summary align="right"><a href="http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html">SimpleBaseline2D (ECCV'2018)</a></summary>
+
+```bibtex
+@inproceedings{xiao2018simple,
+  title={Simple baselines for human pose estimation and tracking},
+  author={Xiao, Bin and Wu, Haiping and Wei, Yichen},
+  booktitle={Proceedings of the European conference on computer vision (ECCV)},
+  pages={466--481},
+  year={2018}
+}
+```
+
+</details>
+
+<!-- [DATASET] -->
+
+<details>
+<summary align="right"><a href="https://openaccess.thecvf.com/content/WACV2021/html/Mathis_Pretraining_Boosts_Out-of-Domain_Robustness_for_Pose_Estimation_WACV_2021_paper.html">Horse-10 (WACV'2021)</a></summary>
+
+```bibtex
+@inproceedings{mathis2021pretraining,
+  title={Pretraining boosts out-of-domain robustness for pose estimation},
+  author={Mathis, Alexander and Biasi, Thomas and Schneider, Steffen and Yuksekgonul, Mert and Rogers, Byron and Bethge, Matthias and Mathis, Mackenzie W},
+  booktitle={Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision},
+  pages={1859--1868},
+  year={2021}
+}
+```
+
+</details>
+
+Results on Horse-10 test set
+
+| Set    |                           Arch                            | Input Size | PCK@0.3 |  NME  |                            ckpt                            |                            log                            |
+| :----- | :-------------------------------------------------------: | :--------: | :-----: | :---: | :--------------------------------------------------------: | :-------------------------------------------------------: |
+| split1 | [pose_resnet_50](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split1.py) |  256x256   |  0.956  | 0.113 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split1-3a3dc37e_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split1_20210405.log.json) |
+| split2 | [pose_resnet_50](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split2.py) |  256x256   |  0.954  | 0.111 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split2-65e2a508_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split2_20210405.log.json) |
+| split3 | [pose_resnet_50](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split3.py) |  256x256   |  0.946  | 0.129 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split3-9637d4eb_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split3_20210405.log.json) |
+| split1 | [pose_resnet_101](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res101_horse10_256x256-split1.py) |  256x256   |  0.958  | 0.115 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split1-1b7c259c_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split1_20210405.log.json) |
+| split2 | [pose_resnet_101](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res101_horse10_256x256-split2.py) |  256x256   |  0.955  | 0.115 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split2-30e2fa87_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split2_20210405.log.json) |
+| split3 | [pose_resnet_101](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res101_horse10_256x256-split3.py) |  256x256   |  0.946  | 0.126 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split3-2eea5bb1_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split3_20210405.log.json) |
+| split1 | [pose_resnet_152](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split1.py) |  256x256   |  0.969  | 0.105 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split1-7e81fe2d_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split1_20210405.log.json) |
+| split2 | [pose_resnet_152](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split2.py) |  256x256   |  0.970  | 0.103 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split2-3b3404a3_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split2_20210405.log.json) |
+| split3 | [pose_resnet_152](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split3.py) |  256x256   |  0.957  | 0.131 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split3-c957dac5_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split3_20210405.log.json) |

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/resnet_horse10.yml

+Collections:
+- Name: SimpleBaseline2D
+  Paper:
+    Title: Simple baselines for human pose estimation and tracking
+    URL: http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html
+  README: https://github.com/open-mmlab/mmpose/blob/master/docs/en/papers/algorithms/simplebaseline2d.md
+Models:
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split1.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: &id001
+    - SimpleBaseline2D
+    Training Data: Horse-10
+  Name: topdown_heatmap_res50_horse10_256x256-split1
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.113
+      PCK@0.3: 0.956
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split1-3a3dc37e_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split2.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res50_horse10_256x256-split2
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.111
+      PCK@0.3: 0.954
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split2-65e2a508_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split3.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res50_horse10_256x256-split3
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.129
+      PCK@0.3: 0.946
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split3-9637d4eb_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res101_horse10_256x256-split1.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res101_horse10_256x256-split1
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.115
+      PCK@0.3: 0.958
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split1-1b7c259c_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res101_horse10_256x256-split2.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res101_horse10_256x256-split2
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.115
+      PCK@0.3: 0.955
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split2-30e2fa87_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res101_horse10_256x256-split3.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res101_horse10_256x256-split3
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.126
+      PCK@0.3: 0.946
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split3-2eea5bb1_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split1.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res152_horse10_256x256-split1
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.105
+      PCK@0.3: 0.969
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split1-7e81fe2d_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split2.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res152_horse10_256x256-split2
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.103
+      PCK@0.3: 0.97
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split2-3b3404a3_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split3.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res152_horse10_256x256-split3
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.131
+      PCK@0.3: 0.957
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split3-c957dac5_20210405.pth

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res101_locust_160x160.py

+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/locust.py'
+]
+evaluation = dict(interval=10, metric=['PCK', 'AUC', 'EPE'], save_best='AUC')
+
+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
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=35,
+    dataset_joints=35,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet101',
+    backbone=dict(type='ResNet', depth=101),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[160, 160],
+    heatmap_size=[40, 40],
+    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'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownGetBboxCenterScale', padding=0.8),
+    dict(type='TopDownRandomShiftBboxCenter', shift_factor=0.25, prob=0.3),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=90, scale_factor=0.3),
+    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='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownGetBboxCenterScale', padding=0.8),
+    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', 'flip_pairs']),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/locust'
+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='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_train.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res152_locust_160x160.py

+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/locust.py'
+]
+evaluation = dict(interval=10, metric=['PCK', 'AUC', 'EPE'], save_best='AUC')
+
+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
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=35,
+    dataset_joints=35,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet152',
+    backbone=dict(type='ResNet', depth=152),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[160, 160],
+    heatmap_size=[40, 40],
+    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'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownGetBboxCenterScale', padding=0.8),
+    dict(type='TopDownRandomShiftBboxCenter', shift_factor=0.25, prob=0.3),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=90, scale_factor=0.3),
+    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='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownGetBboxCenterScale', padding=0.8),
+    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', 'flip_pairs']),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/locust'
+data = dict(
+    samples_per_gpu=32,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_train.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res50_locust_160x160.py

+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/locust.py'
+]
+evaluation = dict(interval=10, metric=['PCK', 'AUC', 'EPE'], save_best='AUC')
+
+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
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=35,
+    dataset_joints=35,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet50',
+    backbone=dict(type='ResNet', depth=50),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[160, 160],
+    heatmap_size=[40, 40],
+    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'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownGetBboxCenterScale', padding=0.8),
+    dict(type='TopDownRandomShiftBboxCenter', shift_factor=0.25, prob=0.3),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=90, scale_factor=0.3),
+    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='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownGetBboxCenterScale', padding=0.8),
+    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', 'flip_pairs']),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/locust'
+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='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_train.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/resnet_locust.md

+<!-- [ALGORITHM] -->
+
+<details>
+<summary align="right"><a href="http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html">SimpleBaseline2D (ECCV'2018)</a></summary>
+
+```bibtex
+@inproceedings{xiao2018simple,
+  title={Simple baselines for human pose estimation and tracking},
+  author={Xiao, Bin and Wu, Haiping and Wei, Yichen},
+  booktitle={Proceedings of the European conference on computer vision (ECCV)},
+  pages={466--481},
+  year={2018}
+}
+```
+
+</details>
+
+<!-- [DATASET] -->
+
+<details>
+<summary align="right"><a href="https://elifesciences.org/articles/47994">Desert Locust (Elife'2019)</a></summary>
+
+```bibtex
+@article{graving2019deepposekit,
+  title={DeepPoseKit, a software toolkit for fast and robust animal pose estimation using deep learning},
+  author={Graving, Jacob M and Chae, Daniel and Naik, Hemal and Li, Liang and Koger, Benjamin and Costelloe, Blair R and Couzin, Iain D},
+  journal={Elife},
+  volume={8},
+  pages={e47994},
+  year={2019},
+  publisher={eLife Sciences Publications Limited}
+}
+```
+
+</details>
+
+Results on Desert Locust test set
+
+| Arch                                                       | Input Size | PCK@0.2 |  AUC  | EPE  |                            ckpt                            |                            log                             |
+| :--------------------------------------------------------- | :--------: | :-----: | :---: | :--: | :--------------------------------------------------------: | :--------------------------------------------------------: |
+| [pose_resnet_50](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res50_locust_160x160.py) |  160x160   |  0.999  | 0.899 | 2.27 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res50_locust_160x160-9efca22b_20210407.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res50_locust_160x160_20210407.log.json) |
+| [pose_resnet_101](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res101_locust_160x160.py) |  160x160   |  0.999  | 0.907 | 2.03 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res101_locust_160x160-d77986b3_20210407.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res101_locust_160x160_20210407.log.json) |
+| [pose_resnet_152](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res152_locust_160x160.py) |  160x160   |  1.000  | 0.926 | 1.48 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res152_locust_160x160-4ea9b372_20210407.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res152_locust_160x160_20210407.log.json) |

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/resnet_locust.yml

+Collections:
+- Name: SimpleBaseline2D
+  Paper:
+    Title: Simple baselines for human pose estimation and tracking
+    URL: http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html
+  README: https://github.com/open-mmlab/mmpose/blob/master/docs/en/papers/algorithms/simplebaseline2d.md
+Models:
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res50_locust_160x160.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: &id001
+    - SimpleBaseline2D
+    Training Data: Desert Locust
+  Name: topdown_heatmap_res50_locust_160x160
+  Results:
+  - Dataset: Desert Locust
+    Metrics:
+      AUC: 0.899
+      EPE: 2.27
+      PCK@0.2: 0.999
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res50_locust_160x160-9efca22b_20210407.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res101_locust_160x160.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Desert Locust
+  Name: topdown_heatmap_res101_locust_160x160
+  Results:
+  - Dataset: Desert Locust
+    Metrics:
+      AUC: 0.907
+      EPE: 2.03
+      PCK@0.2: 0.999
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res101_locust_160x160-d77986b3_20210407.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res152_locust_160x160.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Desert Locust
+  Name: topdown_heatmap_res152_locust_160x160
+  Results:
+  - Dataset: Desert Locust
+    Metrics:
+      AUC: 0.926
+      EPE: 1.48
+      PCK@0.2: 1.0
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res152_locust_160x160-4ea9b372_20210407.pth

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/hrnet_macaque.md

+<!-- [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://www.ncbi.nlm.nih.gov/pmc/articles/pmc7874091/">MacaquePose (bioRxiv'2020)</a></summary>
+
+```bibtex
+@article{labuguen2020macaquepose,
+  title={MacaquePose: A novel ‘in the wild’macaque monkey pose dataset for markerless motion capture},
+  author={Labuguen, Rollyn and Matsumoto, Jumpei and Negrete, Salvador and Nishimaru, Hiroshi and Nishijo, Hisao and Takada, Masahiko and Go, Yasuhiro and Inoue, Ken-ichi and Shibata, Tomohiro},
+  journal={bioRxiv},
+  year={2020},
+  publisher={Cold Spring Harbor Laboratory}
+}
+```
+
+</details>
+
+Results on MacaquePose with ground-truth detection 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/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/hrnet_w32_macaque_256x192.py) |  256x192   | 0.814 |      0.953      |      0.918      | 0.851 |      0.969      | [ckpt](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_macaque_256x192-f7e9e04f_20210407.pth) | [log](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_macaque_256x192_20210407.log.json) |
+| [pose_hrnet_w48](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/hrnet_w48_macaque_256x192.py) |  256x192   | 0.818 |      0.963      |      0.917      | 0.855 |      0.971      | [ckpt](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_macaque_256x192-9b34b02a_20210407.pth) | [log](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_macaque_256x192_20210407.log.json) |

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/hrnet_macaque.yml

+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/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/hrnet_w32_macaque_256x192.py
+  In Collection: HRNet
+  Metadata:
+    Architecture: &id001
+    - HRNet
+    Training Data: MacaquePose
+  Name: topdown_heatmap_hrnet_w32_macaque_256x192
+  Results:
+  - Dataset: MacaquePose
+    Metrics:
+      AP: 0.814
+      AP@0.5: 0.953
+      AP@0.75: 0.918
+      AR: 0.851
+      AR@0.5: 0.969
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_macaque_256x192-f7e9e04f_20210407.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/hrnet_w48_macaque_256x192.py
+  In Collection: HRNet
+  Metadata:
+    Architecture: *id001
+    Training Data: MacaquePose
+  Name: topdown_heatmap_hrnet_w48_macaque_256x192
+  Results:
+  - Dataset: MacaquePose
+    Metrics:
+      AP: 0.818
+      AP@0.5: 0.963
+      AP@0.75: 0.917
+      AR: 0.855
+      AR@0.5: 0.971
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_macaque_256x192-9b34b02a_20210407.pth

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/hrnet_w32_macaque_256x192.py

+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/macaque.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
+log_config = dict(
+    interval=10,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+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='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))),
+    ),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=32,
+        out_channels=channel_cfg['num_output_channels'],
+        num_deconv_layers=0,
+        extra=dict(final_conv_kernel=1, ),
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+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=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+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='TopDownGenerateTarget', sigma=2),
+    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/macaque'
+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='AnimalMacaqueDataset',
+        ann_file=f'{data_root}/annotations/macaque_train.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalMacaqueDataset',
+        ann_file=f'{data_root}/annotations/macaque_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalMacaqueDataset',
+        ann_file=f'{data_root}/annotations/macaque_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/hrnet_w48_macaque_256x192.py

+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/macaque.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
+log_config = dict(
+    interval=10,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+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='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='TopdownHeatmapSimpleHead',
+        in_channels=48,
+        out_channels=channel_cfg['num_output_channels'],
+        num_deconv_layers=0,
+        extra=dict(final_conv_kernel=1, ),
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+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=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+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='TopDownGenerateTarget', sigma=2),
+    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/macaque'
+data = dict(
+    samples_per_gpu=32,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalMacaqueDataset',
+        ann_file=f'{data_root}/annotations/macaque_train.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalMacaqueDataset',
+        ann_file=f'{data_root}/annotations/macaque_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalMacaqueDataset',
+        ann_file=f'{data_root}/annotations/macaque_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/res101_macaque_256x192.py

+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/macaque.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
+log_config = dict(
+    interval=10,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+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),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+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=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+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='TopDownGenerateTarget', sigma=2),
+    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/macaque'
+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='AnimalMacaqueDataset',
+        ann_file=f'{data_root}/annotations/macaque_train.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalMacaqueDataset',
+        ann_file=f'{data_root}/annotations/macaque_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalMacaqueDataset',
+        ann_file=f'{data_root}/annotations/macaque_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/res152_macaque_256x192.py

+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/macaque.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
+log_config = dict(
+    interval=10,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+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),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+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=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+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='TopDownGenerateTarget', sigma=2),
+    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/macaque'
+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='AnimalMacaqueDataset',
+        ann_file=f'{data_root}/annotations/macaque_train.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalMacaqueDataset',
+        ann_file=f'{data_root}/annotations/macaque_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalMacaqueDataset',
+        ann_file=f'{data_root}/annotations/macaque_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/res50_macaque_256x192.py

+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/macaque.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
+log_config = dict(
+    interval=10,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+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),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+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=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+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='TopDownGenerateTarget', sigma=2),
+    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/macaque'
+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='AnimalMacaqueDataset',
+        ann_file=f'{data_root}/annotations/macaque_train.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalMacaqueDataset',
+        ann_file=f'{data_root}/annotations/macaque_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalMacaqueDataset',
+        ann_file=f'{data_root}/annotations/macaque_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/resnet_macaque.md

+<!-- [ALGORITHM] -->
+
+<details>
+<summary align="right"><a href="http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html">SimpleBaseline2D (ECCV'2018)</a></summary>
+
+```bibtex
+@inproceedings{xiao2018simple,
+  title={Simple baselines for human pose estimation and tracking},
+  author={Xiao, Bin and Wu, Haiping and Wei, Yichen},
+  booktitle={Proceedings of the European conference on computer vision (ECCV)},
+  pages={466--481},
+  year={2018}
+}
+```
+
+</details>
+
+<!-- [DATASET] -->
+
+<details>
+<summary align="right"><a href="https://www.ncbi.nlm.nih.gov/pmc/articles/pmc7874091/">MacaquePose (bioRxiv'2020)</a></summary>
+
+```bibtex
+@article{labuguen2020macaquepose,
+  title={MacaquePose: A novel ‘in the wild’macaque monkey pose dataset for markerless motion capture},
+  author={Labuguen, Rollyn and Matsumoto, Jumpei and Negrete, Salvador and Nishimaru, Hiroshi and Nishijo, Hisao and Takada, Masahiko and Go, Yasuhiro and Inoue, Ken-ichi and Shibata, Tomohiro},
+  journal={bioRxiv},
+  year={2020},
+  publisher={Cold Spring Harbor Laboratory}
+}
+```
+
+</details>
+
+Results on MacaquePose with ground-truth detection bounding boxes
+
+| Arch                                          | Input Size |  AP   | AP<sup>50</sup> | AP<sup>75</sup> |  AR   | AR<sup>50</sup> |                     ckpt                      |                      log                      |
+| :-------------------------------------------- | :--------: | :---: | :-------------: | :-------------: | :---: | :-------------: | :-------------------------------------------: | :-------------------------------------------: |
+| [pose_resnet_50](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/res50_macaque_256x192.py) |  256x192   | 0.799 |      0.952      |      0.919      | 0.837 |      0.964      | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res50_macaque_256x192-98f1dd3a_20210407.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res50_macaque_256x192_20210407.log.json) |
+| [pose_resnet_101](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/res101_macaque_256x192.py) |  256x192   | 0.790 |      0.953      |      0.908      | 0.828 |      0.967      | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res101_macaque_256x192-e3b9c6bb_20210407.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res101_macaque_256x192_20210407.log.json) |
+| [pose_resnet_152](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/macaque/res152_macaque_256x192.py) |  256x192   | 0.794 |      0.951      |      0.915      | 0.834 |      0.968      | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res152_macaque_256x192-c42abc02_20210407.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res152_macaque_256x192_20210407.log.json) |