LSPe - Leeds Sports Pose Extended

The Leeds Sports Pose extended dataset contains 10,000 images gathered from Flickr searches for the tags ‘parkour’, ‘gymnastics’, and ‘athletics’ and consists of poses deemed to be challenging to estimate. Each image has a corresponding annotation gathered from Amazon Mechanical Turk and as such cannot be guaranteed to be highly accurate. The images have been scaled such that the annotated person is roughly 150 pixels in length. Each image has been annotated with up to 14 visible joint locations.

Use cases

Human body joint detection.

Properties

  • name: leeds_sports_pose_extended
  • keywords: image_processing, detection, human_pose, keypoints
  • dataset size: 206,2 MB
  • is downloadable: yes
  • tasks:
    • keypoints: (default)
      • primary use: human body joint detection
      • description: Contains image files and body parts keypoint coordinates for detecting human body joints in images
      • sets: train, test
      • metadata file size in disk: 473,7 kB
      • has annotations: yes
        • which:
          • body joint keypoints

Note

This dataset is essentially the same as the leeds_sports_pose but contains more training samples.

Metadata structure (HDF5)

Task: keypoints

/
├── train/
│   ├── image_filenames   # dtype=np.uint8, shape=(11000,104)  (note: string in ASCII format)
│   ├── keypoint_names    # dtype=np.uint8, shape=(14,15)    (note: string in ASCII format)
│   ├── keypoints         # dtype=np.float, shape=(11000,14,3)
│   ├── object_fields     # dtype=np.uint8, shape=(2,16)     (note: string in ASCII format)
│   └── object_ids        # dtype=np.int32, shape=(11000,2)
│
└── test/
    ├── image_filenames   # dtype=np.uint8, shape=(1000,104)  (note: string in ASCII format)
    ├── keypoint_names    # dtype=np.uint8, shape=(14,15)    (note: string in ASCII format)
    ├── keypoints         # dtype=np.float, shape=(1000,14,3)
    ├── object_fields     # dtype=np.uint8, shape=(2,16)     (note: string in ASCII format)
    └── object_ids        # dtype=np.int32, shape=(1000,2)

Fields

  • image_filenames: image file path+name
    • available in: train, test
    • dtype: np.uint8
    • is padded: True
    • fill value: 0
    • note: strings stored in ASCII format
  • keypoint_names: body joint names
    • available in: train, test
    • dtype: np.uint8
    • is padded: True
    • fill value: 0
    • note: strings stored in ASCII format
  • keypoints: keypoint coordinates
    • available in: train, test
    • dtype: np.float
    • is padded: False
    • fill value: -1
    • note: keypoint format [x1,y1,is_visible]
  • object_fields: list of field names of the object id list
    • available in: train, test
    • dtype: np.uint8
    • is padded: True
    • fill value: 0
    • note: strings stored in ASCII format
    • note: key field (field name aggregator)
  • object_ids: list of field ids
    • available in: train, test
    • dtype: np.int32
    • is padded: False
    • fill value: -1
    • note: key field (field id aggregator)

Disclaimer

All rights reserved to the original creators of Leeds Sports Pose extended.

For information about the dataset and its terms of use, please see this link.