• Description:

The dataset contains 7200 color images of 100 objects (72 images per object). The objects have a wide variety of complex geometric and reflectance characteristics. The objects were placed on a motorized turntable against a black background. The turntable was rotated through 360 degrees to vary object pose with respect to a fxed color camera. Images of the objects were taken at pose intervals of 5 degrees.This corresponds to 72 poses per object

Split Examples
'train' 7,200
  • Features:
    'angle': tf.int64,
    'angle_label': ClassLabel(shape=(), dtype=tf.int64, num_classes=72),
    'image': Image(shape=(128, 128, 3), dtype=tf.uint8),
    'object_id': ClassLabel(shape=(), dtype=tf.int64, num_classes=100),
  title={Columbia object image library (coil-20)},
  author={Nene, Sameer A and Nayar, Shree K and Murase, Hiroshi and others},
  publisher={Technical report CUCS-005-96}