Mikado Dataset

A Synthetic Dataset of Dense Homogeneous Object Layouts
for Occlusion-aware Instance Segmentation

Samples from Mikado

Collaborators

Matthieu Grard, Siléane & Ecole Centrale de Lyon, LIRIS, France
Emmanuel Dellandréa, Ecole Centrale de Lyon, LIRIS, France
Liming Chen, Ecole Centrale de Lyon, LIRIS, France

Citation

If you use the Mikado dataset in your research, please cite the following paper:

M. Grard, E. Dellandrea, and L. Chen, "Deep Multicameral Decoding for Localizing Unoccluded Object Instances from a Single RGB Image" in International Journal of Computer Vision (IJCV), 2020. DOI: https://doi.org/10.1007/s11263-020-01323-0