hyperpose.Dataset.mscoco_dataset package¶
Submodules¶
hyperpose.Dataset.mscoco_dataset.dataset module¶
-
class
hyperpose.Dataset.mscoco_dataset.dataset.
MSCOCO_dataset
(config, input_kpt_cvter=None, output_kpt_cvter=None, dataset_filter=None)¶ Bases:
hyperpose.Dataset.base_dataset.Base_dataset
a dataset class specified for coco dataset, provides uniform APIs
Methods
get_eval_dataset
(self[, in_list])provide uniform tensorflow dataset for evaluating
get_train_dataset
(self[, in_list, …])provide uniform tensorflow dataset for training
official_eval
(self, pd_anns[, eval_dir])providing official evaluation of COCO dataset
prepare_dataset
(self)download,extract, and reformat the dataset the official format is in zip format, extract it into json files and image files.
visualize
(self, vis_num)visualize annotations of the train dataset
generate_eval_data
generate_test_data
generate_train_data
get_colors
get_dataset_type
get_eval_datasize
get_input_kpt_cvter
get_output_kpt_cvter
get_parts
get_test_dataset
get_test_datasize
get_train_datasize
official_test
set_dataset_version
set_input_kpt_cvter
set_output_kpt_cvter
-
generate_eval_data
(self)¶
-
generate_test_data
(self)¶
-
generate_train_data
(self)¶
-
get_colors
(self)¶
-
get_dataset_type
(self)¶
-
get_input_kpt_cvter
(self)¶
-
get_output_kpt_cvter
(self)¶
-
get_parts
(self)¶
-
official_eval
(self, pd_anns, eval_dir='./eval_dir')¶ providing official evaluation of COCO dataset
using pycocotool.cocoeval class to perform official evaluation. output model metrics of MAPs on coco evaluation dataset
- Parameters
- arg1String
A string path of the json file in the same format of cocoeval annotation file(person_keypoints_val2017.json) which contains predicted results. one can refer the evaluation pipeline of models for generation procedure of this json file.
- arg2String
A string path indicates where the json files of filtered intersection part of predict results and ground truth the filtered prediction file is stored in eval_dir/pd_ann.json the filtered ground truth file is stored in eval_dir/gt_ann.json
- Returns
- None
-
official_test
(self, pd_anns, test_dir='./test_dir')¶
-
prepare_dataset
(self)¶ download,extract, and reformat the dataset the official format is in zip format, extract it into json files and image files.
- Parameters
- None
- Returns
- None
-
set_input_kpt_cvter
(self, input_kpt_cvter)¶
-
set_output_kpt_cvter
(self, output_kpt_cvter)¶
-
visualize
(self, vis_num)¶ visualize annotations of the train dataset
visualize the annotation points in the image to help understand and check annotation the visualized image will be saved in the “data_vis_dir” of the corresponding model directory(specified by model name). the visualized annotations are from the train dataset.
- Parameters
- arg1Int
An integer indicates how many images with their annotations are going to be visualized.
- Returns
- None
-
-
hyperpose.Dataset.mscoco_dataset.dataset.
init_dataset
(config)¶
hyperpose.Dataset.mscoco_dataset.define module¶
-
class
hyperpose.Dataset.mscoco_dataset.define.
CocoPart
(value)¶ Bases:
enum.Enum
An enumeration.
-
LAnkle
= 15¶
-
LEar
= 3¶
-
LElbow
= 7¶
-
LHip
= 11¶
-
LKnee
= 13¶
-
LShoulder
= 5¶
-
LWrist
= 9¶
-
Leye
= 1¶
-
Nose
= 0¶
-
RAnkle
= 16¶
-
REar
= 4¶
-
RElbow
= 8¶
-
RHip
= 12¶
-
RKnee
= 14¶
-
RShoulder
= 6¶
-
RWrist
= 10¶
-
Reye
= 2¶
-
-
hyperpose.Dataset.mscoco_dataset.define.
opps_input_converter
(coco_kpts)¶
-
hyperpose.Dataset.mscoco_dataset.define.
opps_output_converter
(kpt_list)¶
-
hyperpose.Dataset.mscoco_dataset.define.
pifpaf_input_converter
(coco_kpts)¶
-
hyperpose.Dataset.mscoco_dataset.define.
pifpaf_output_converter
(kpt_list)¶
-
hyperpose.Dataset.mscoco_dataset.define.
ppn_input_converter
(coco_kpts)¶
-
hyperpose.Dataset.mscoco_dataset.define.
ppn_output_converter
(kpt_list)¶
hyperpose.Dataset.mscoco_dataset.format module¶
-
class
hyperpose.Dataset.mscoco_dataset.format.
CocoMeta
(image_id, img_url, img_meta, kpts_infos, masks, bbxs, is_crowd)¶ Bases:
object
Be used in PoseInfo.
-
class
hyperpose.Dataset.mscoco_dataset.format.
PoseInfo
(image_base_dir, anno_path, with_mask=True, dataset_filter=None, eval=False)¶ Bases:
object
Use COCO for pose estimation, returns images with people only.
Methods
get_image_annos
(self)Read JSON file, and get and check the image list.
get_bbx_list
get_bbxs
get_image_id_list
get_image_list
get_keypoints
get_kpt_list
get_mask_list
load_images
-
get_bbx_list
(self)¶
-
static
get_bbxs
(annos_info)¶
-
get_image_annos
(self)¶ Read JSON file, and get and check the image list. Skip missing images.
-
get_image_id_list
(self)¶
-
get_image_list
(self)¶
-
static
get_keypoints
(annos_info)¶
-
get_kpt_list
(self)¶
-
get_mask_list
(self)¶
-
load_images
(self)¶
-
hyperpose.Dataset.mscoco_dataset.generate module¶
-
hyperpose.Dataset.mscoco_dataset.generate.
generate_eval_data
(val_imgs_path, val_anns_path, dataset_filter=None)¶
-
hyperpose.Dataset.mscoco_dataset.generate.
generate_test_data
(test_imgs_path, test_anns_path)¶
-
hyperpose.Dataset.mscoco_dataset.generate.
generate_train_data
(train_imgs_path, train_anns_path, dataset_filter=None, input_kpt_cvter=<function <lambda> at 0x7f0a853361e0>)¶
hyperpose.Dataset.mscoco_dataset.prepare module¶
-
hyperpose.Dataset.mscoco_dataset.prepare.
prepare_dataset
(data_path='./data', version='2017', task='person')¶ Download MSCOCO Dataset. Both 2014 and 2017 dataset have train, validate and test sets, but 2017 version put less data into the validation set (115k train, 5k validate) i.e. has more training data.
- Parameters
- pathstr
The path that the data is downloaded to, defaults is
data/mscoco...
.- datasetstr
The MSCOCO dataset version, 2014 or 2017.
- taskstr
person for pose estimation, caption for image captioning, instance for segmentation.
- Returns
- train_image_pathstr
Folder path of all training images.
- train_ann_pathstr
File path of training annotations.
- val_image_pathstr
Folder path of all validating images.
- val_ann_pathstr
File path of validating annotations.
- test_image_pathstr
Folder path of all testing images.
- test_ann_pathNone
File path of testing annotations, but as the test sets of MSCOCO 2014 and 2017 do not have annotation, returns None.
References
Examples
>>> train_im_path, train_ann_path, val_im_path, val_ann_path, _, _ = ... tl.files.load_mscoco_dataset('data', '2017')