Fold and Occlusion Boundary Detection


1. Task

Given a single RGB image, perform joint boundary detection and occlusion-versus-fold classification: deciding whether a pixel is a boundary (fold or occlusion) and if so, which kind it is.

2. Evaluation Metrics

For a boundary pixel to be considered correct, it has to be labeled correctly as either occlusion or fold.

For a more detailed explanation of the above metrics, please refer to the paper.

3. Leaderboard

Method OIS ODS AP
Hourglass (OASIS)[1] 0.639 0.581 0.530

References

[1]
Hourglass (OASIS)
Chen, W., Qian, S., Fan, D., Kojima, N., Hamilton, M., & Deng, J. OASIS: A Large-Scale Dataset for Single Image 3D in the Wild. CVPR 2020.