easycv.datasets.pose.data_sources package¶
- class easycv.datasets.pose.data_sources.PoseTopDownSourceCoco(ann_file, img_prefix, data_cfg, dataset_info=None, test_mode=False)[source]¶
Bases:
easycv.datasets.pose.data_sources.top_down.PoseTopDownSource
CocoSource for top-down pose estimation.
Microsoft COCO: Common Objects in Context’ ECCV’2014 More details can be found in the `paper .
The source loads raw features to build a data meta object containing the image info, annotation info and others.
COCO keypoint indexes:
0: 'nose', 1: 'left_eye', 2: 'right_eye', 3: 'left_ear', 4: 'right_ear', 5: 'left_shoulder', 6: 'right_shoulder', 7: 'left_elbow', 8: 'right_elbow', 9: 'left_wrist', 10: 'right_wrist', 11: 'left_hip', 12: 'right_hip', 13: 'left_knee', 14: 'right_knee', 15: 'left_ankle', 16: 'right_ankle'
- Parameters
ann_file (str) – Path to the annotation file.
img_prefix (str) – Path to a directory where images are held. Default: None.
data_cfg (dict) – config
dataset_info (DatasetInfo) – A class containing all dataset info.
test_mode (bool) – Store True when building test or
dataset. Default (validation) – False.
- class easycv.datasets.pose.data_sources.PoseTopDownSource(ann_file, img_prefix, data_cfg, dataset_info, coco_style=True, test_mode=False)[source]¶
Bases:
object
Class for keypoint 2D top-down pose estimation with single-view RGB image as the data source.
- Parameters
ann_file (str) – Path to the annotation file.
img_prefix (str) – Path to a directory where images are held. Default: None.
data_cfg (dict) – config
dataset_info (DatasetInfo) – A class containing all dataset info.
coco_style (bool) – Whether the annotation json is coco-style. Default: True
test_mode (bool) – Store True when building test or validation dataset. Default: False.
- class easycv.datasets.pose.data_sources.HandCocoPoseTopDownSource(ann_file, img_prefix, data_cfg, dataset_info=None, test_mode=False)[source]¶
Bases:
easycv.datasets.pose.data_sources.top_down.PoseTopDownSource
Coco Whole-Body-Hand Source for top-down hand pose estimation.
“Whole-Body Human Pose Estimation in the Wild”, ECCV’2020. More details can be found in the paper .
The dataset loads raw features and apply specified transforms to return a dict containing the image tensors and other information.
COCO-WholeBody Hand keypoint indexes:
0: 'wrist', 1: 'thumb1', 2: 'thumb2', 3: 'thumb3', 4: 'thumb4', 5: 'forefinger1', 6: 'forefinger2', 7: 'forefinger3', 8: 'forefinger4', 9: 'middle_finger1', 10: 'middle_finger2', 11: 'middle_finger3', 12: 'middle_finger4', 13: 'ring_finger1', 14: 'ring_finger2', 15: 'ring_finger3', 16: 'ring_finger4', 17: 'pinky_finger1', 18: 'pinky_finger2', 19: 'pinky_finger3', 20: 'pinky_finger4'
- Parameters
ann_file (str) – Path to the annotation file.
img_prefix (str) – Path to a directory where images are held. Default: None.
data_cfg (dict) – config
dataset_info (DatasetInfo) – A class containing all dataset info.
test_mode (bool) – Store True when building test or validation dataset. Default: False.
- class easycv.datasets.pose.data_sources.WholeBodyCocoTopDownSource(ann_file, img_prefix, data_cfg, dataset_info=None, test_mode=False)[source]¶
Bases:
easycv.datasets.pose.data_sources.top_down.PoseTopDownSource
CocoWholeBodyDataset dataset for top-down pose estimation.
“Whole-Body Human Pose Estimation in the Wild”, ECCV’2020. More details can be found in the paper .
The dataset loads raw features and apply specified transforms to return a dict containing the image tensors and other information.
COCO-WholeBody keypoint indexes:
0-16: 17 body keypoints, 17-22: 6 foot keypoints, 23-90: 68 face keypoints, 91-132: 42 hand keypoints In total, we have 133 keypoints for wholebody pose estimation.
- Parameters
ann_file (str) – Path to the annotation file.
img_prefix (str) – Path to a directory where images are held. Default: None.
data_cfg (dict) – config
test_mode (bool) – Store True when building test or validation dataset. Default: False.
- class easycv.datasets.pose.data_sources.PoseTopDownSourceCoco2017(data_cfg, path='', download=True, split='train', dataset_info=None, test_mode=False)[source]¶
Bases:
easycv.datasets.pose.data_sources.coco.PoseTopDownSourceCoco
- Parameters
path – target dir
download – whether download
split – train or val
data_cfg (dict) – config
dataset_info (DatasetInfo) – A class containing all dataset info.
test_mode (bool) – Store True when building test or
dataset. Default (validation) – False.
- class easycv.datasets.pose.data_sources.PoseTopDownSourceCrowdPose(ann_file, img_prefix, data_cfg, dataset_info=None, test_mode=False, **kwargs)[source]¶
Bases:
easycv.datasets.pose.data_sources.top_down.PoseTopDownSource
CrowdPose keypoint indexes:
0 ‘left_shoulder’, 1 ‘right_shoulder’, 2 ‘left_elbow’, 3 ‘right_elbow’, 4 ‘left_wrist’, 5 ‘right_wrist’, 6 ‘left_hip’, 7 ‘right_hip’, 8 ‘left_knee’, 9 ‘right_knee’, 10 ‘left_ankle’, 11 ‘right_ankle’, 12 ‘head’, 13 ‘neck’
- Parameters
ann_file (str) – Path to the annotation file.
img_prefix (str) – Path to a directory where images are held. Default: None.
data_cfg (dict) – config
dataset_info (DatasetInfo) – A class containing all dataset info.
test_mode (bool) – Store True when building test or
dataset. Default (validation) – False.
- class easycv.datasets.pose.data_sources.PoseTopDownSourceChHuman(ann_file, img_prefix, data_cfg, subset=None, dataset_info=None, test_mode=False, **kwargs)[source]¶
Bases:
easycv.datasets.pose.data_sources.top_down.PoseTopDownSource
Oc Human Source for top-down pose estimation.
The source loads raw features to build a data meta object containing the image info, annotation info and others.
Oc Human keypoint indexes:
0: 'nose', 1: 'left_eye', 2: 'right_eye', 3: 'left_ear', 4: 'right_ear', 5: 'left_shoulder', 6: 'right_shoulder', 7: 'left_elbow', 8: 'right_elbow', 9: 'left_wrist', 10: 'right_wrist', 11: 'left_hip', 12: 'right_hip', 13: 'left_knee', 14: 'right_knee', 15: 'left_ankle', 16: 'right_ankle'
- Parameters
ann_file (str) – Path to the annotation file.
img_prefix (str) – Path to a directory where images are held. Default: None.
data_cfg (dict) – config
subset – Applicable to non-coco or coco style data sets, if subset == train or val or test, in non-coco style else subset == None , in coco style
dataset_info (DatasetInfo) – A class containing all dataset info.
test_mode (bool) – Store True when building test or
- class easycv.datasets.pose.data_sources.PoseTopDownSourceMpii(data_cfg, path='/home/docs/.cache/easycv/', download=False, dataset_info=None, test_mode=False, **kwargs)[source]¶
Bases:
easycv.datasets.pose.data_sources.top_down.PoseTopDownSource
Oc Human Source for top-down pose estimation.
The source loads raw features to build a data meta object containing the image info, annotation info and others.
Oc Human keypoint indexes:
0: 'right_ankle', 1: 'right_knee', 2: 'right_hip', 3: 'left_hip', 4: 'right_ear', 5: 'left_ankle', 6: 'pelvis', 7: 'thorax', 8: 'neck', 9: 'head', 10: 'right_wrist', 11: 'right_elbow', 12: 'right_shoulder', 13: 'left_shoulder', 14: 'left_elbow', 15: 'left_wrist'
- Parameters
data_cfg (dict) – config
path – This parameter is optional. If download is True and path is not provided, a temporary directory is automatically created for downloading
download – If the value is True, the file is automatically downloaded to the path directory. If False, automatic download is not supported and data in the path is used
dataset_info (DatasetInfo) – A class containing all dataset info.
test_mode (bool) – Store True when building test or
Submodules¶
easycv.datasets.pose.data_sources.coco module¶
- class easycv.datasets.pose.data_sources.coco.PoseTopDownSourceCoco(ann_file, img_prefix, data_cfg, dataset_info=None, test_mode=False)[source]¶
Bases:
easycv.datasets.pose.data_sources.top_down.PoseTopDownSource
CocoSource for top-down pose estimation.
Microsoft COCO: Common Objects in Context’ ECCV’2014 More details can be found in the `paper .
The source loads raw features to build a data meta object containing the image info, annotation info and others.
COCO keypoint indexes:
0: 'nose', 1: 'left_eye', 2: 'right_eye', 3: 'left_ear', 4: 'right_ear', 5: 'left_shoulder', 6: 'right_shoulder', 7: 'left_elbow', 8: 'right_elbow', 9: 'left_wrist', 10: 'right_wrist', 11: 'left_hip', 12: 'right_hip', 13: 'left_knee', 14: 'right_knee', 15: 'left_ankle', 16: 'right_ankle'
- Parameters
ann_file (str) – Path to the annotation file.
img_prefix (str) – Path to a directory where images are held. Default: None.
data_cfg (dict) – config
dataset_info (DatasetInfo) – A class containing all dataset info.
test_mode (bool) – Store True when building test or
dataset. Default (validation) – False.
- class easycv.datasets.pose.data_sources.coco.PoseTopDownSourceCoco2017(data_cfg, path='', download=True, split='train', dataset_info=None, test_mode=False)[source]¶
Bases:
easycv.datasets.pose.data_sources.coco.PoseTopDownSourceCoco
- Parameters
path – target dir
download – whether download
split – train or val
data_cfg (dict) – config
dataset_info (DatasetInfo) – A class containing all dataset info.
test_mode (bool) – Store True when building test or
dataset. Default (validation) – False.
easycv.datasets.pose.data_sources.top_down module¶
- class easycv.datasets.pose.data_sources.top_down.PoseTopDownSource(ann_file, img_prefix, data_cfg, dataset_info, coco_style=True, test_mode=False)[source]¶
Bases:
object
Class for keypoint 2D top-down pose estimation with single-view RGB image as the data source.
- Parameters
ann_file (str) – Path to the annotation file.
img_prefix (str) – Path to a directory where images are held. Default: None.
data_cfg (dict) – config
dataset_info (DatasetInfo) – A class containing all dataset info.
coco_style (bool) – Whether the annotation json is coco-style. Default: True
test_mode (bool) – Store True when building test or validation dataset. Default: False.