Help protect the Great Barrier Reef with TensorFlow on Kaggle Join Challenge


LVIS: A dataset for large vocabulary instance segmentation.

Split Examples
'test' 19,822
'train' 100,170
'validation' 19,809
  • Features:
    'image': Image(shape=(None, None, 3), dtype=tf.uint8),
    'image/id': tf.int64,
    'neg_category_ids': Sequence(ClassLabel(shape=(), dtype=tf.int64, num_classes=1203)),
    'not_exhaustive_category_ids': Sequence(ClassLabel(shape=(), dtype=tf.int64, num_classes=1203)),
    'objects': Sequence({
        'area': tf.int64,
        'bbox': BBoxFeature(shape=(4,), dtype=tf.float32),
        'id': tf.int64,
        'label': ClassLabel(shape=(), dtype=tf.int64, num_classes=1203),
        'segmentation': Image(shape=(None, None, 1), dtype=tf.uint8),


  • Citation:
  title={ {LVIS}: A Dataset for Large Vocabulary Instance Segmentation},
  author={Gupta, Agrim and Dollar, Piotr and Girshick, Ross},
  booktitle={Proceedings of the {IEEE} Conference on Computer Vision and Pattern Recognition},