license: apache-2.0
task_categories:
- image-feature-extraction
- image-classification
- image-to-3d
- image-segmentation
size_categories:
- 1M<n<10M
Dataset Details
This dataset derives from Coil100. There are more than 1,1M images of 100 objects. Each object was turned on a turnable through 360 degrees to vary object pose with respect to a fixed color camera. Images of the objects were taken at pose intervals of 5 degrees. This corresponds to 72 poses per object.
In addition to the original dataset, planar rotation (9 angles) and 18 scaling factors have been applied so that there are no dependencies between factors. Objects have a wide variety of complex geometric and reflectance characteristics.
This augmented version of Coil100 has been designed especially for Disentangled Representation Learning for real images, the Factors of Variations are:
| Factors | # values |
|---|---|
| Object | 100 |
| 3D Pose | 72 |
| Rotation | 9 |
| Scale | 18 |
The binarized version is also available.
How to download
With Python > 3.0 install
pip install huggingface_hub
Then to download the RGB dataset run
python download_coil100.py
And if you want the binary version run
python download_coil100_binary.py
Citation
if you use the dataset, please cite us:
BibTeX:
@article{dapueto2024transferring,
title={Transferring disentangled representations: bridging the gap between synthetic and real images},
author={Dapueto, Jacopo and Noceti, Nicoletta and Odone, Francesca},
journal={arXiv preprint arXiv:2409.18017},
year={2024}
}
Uses
This dataset is intended for non-commercial research purposes only.