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.
For a boundary pixel to be considered correct, it has to be labeled correctly as either occlusion or fold.
OIS: The F-score by optimal threshold per image.
ODS: The F-score by a fixed threshold.
AP: The average precision.
For a more detailed explanation of the above metrics, please refer to the paper.
Method | OIS | ODS | AP |
---|---|---|---|
Hourglass (OASIS)[1] | 0.639 | 0.581 | 0.530 |