FLIC - Frames Labeled In Cinema

Collection of 5003 images automatically from popular Hollywood movies. It contains annotations of upper body joints only. It has 3987 images for training and 1016 for testing.

Use cases

Human body joint detection.

Properties

  • name: flic
  • keywords: image_processing, detection, human_pose, keypoints
  • dataset size: 300,3 MB
  • is downloadable: yes
  • tasks:
    • keypoints: (default)
      • primary use: image classification
      • description: Contains keypoint coordinates (upper body joints only) and bounding boxes of the torso for body joint detection.
      • sets: train, test
      • metadata file size: 582,0 kB
      • has annotations: yes
        • which:
          • image filenames
          • upper body joint coordinates
          • torso bounding box

Metadata structure (HDF5)

Task: classification

/
├── train/
│   ├── image_filenames   # dtype=np.uint8, shape=(3987,108)  (note: string in ASCII format)
│   ├── movienames        # dtype=np.uint8, shape=(3987,31)   (note: string in ASCII format)
│   ├── width             # dtype=np.int32, shape=(3987,)
│   ├── height            # dtype=np.int32, shape=(3987,)
│   ├── torso_boxes       # dtype=np.float, shape=(3987,4)
│   ├── keypoints         # dtype=np.float, shape=(3987,11,3)
│   ├── keypoint_names    # dtype=np.uint8, shape=(11,15)    (note: string in ASCII format)
│   ├── object_fields     # dtype=np.uint8, shape=(5,16)     (note: string in ASCII format)
│   └── object_ids        # dtype=np.int32, shape=(3987,5)
│
└── test/
    ├── image_filenames   # dtype=np.uint8, shape=(1016,108)  (note: string in ASCII format)
    ├── movienames        # dtype=np.uint8, shape=(1016,31)   (note: string in ASCII format)
    ├── width             # dtype=np.int32, shape=(1016,)
    ├── height            # dtype=np.int32, shape=(1016,)
    ├── torso_boxes       # dtype=np.float, shape=(1016,4)
    ├── keypoints         # dtype=np.float, shape=(1016,11,3)
    ├── keypoint_names    # dtype=np.uint8, shape=(11,15)    (note: string in ASCII format)
    ├── object_fields     # dtype=np.uint8, shape=(5,16)     (note: string in ASCII format)
    └── object_ids        # dtype=np.int32, shape=(1016,5)

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
  • movienames: name of the movie where the image was taken from
    • available in: train, test
    • dtype: np.uint8
    • is padded: True
    • fill value: 0
    • note: strings stored in ASCII format
  • width: image width
    • available in: train, test
    • dtype: np.int32
    • is padded: False
    • fill value: -1
  • height: image height
    • available in: train, test
    • dtype: np.int32
    • is padded: False
    • fill value: -1
  • torso_boxes: torso bounding box
    • available in: train, test
    • dtype: np.float
    • is padded: False
    • fill value: -1
    • note: bbox format [x1,y1,x2,y2]
  • keypoints: body joint coordinates
    • available in: train, test
    • dtype: np.float
    • is padded: False
    • fill value: -1
    • note: keypoint format [x1,y1,is_visible]
  • keypoint_names: body joint name
    • available in: train, test
    • dtype: np.uint8
    • is padded: True
    • fill value: 0
    • note: strings stored in ASCII format
  • 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 Frames Labeled In Cinema.

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