Table 1. Evaluation on BSDS500. Higher is better for all measures except VI, for which lower is better.
Algorithm | Covering | RI | VI | F-measure | ||||||
ODS | OIS | Best | ODS | OIS | ODS | OIS | ODS | OIS | AP | |
human | 0.72 | 0.72 | – | 0.88 | 0.88 | 1.17 | 1.17 | 0.80 | 0.80 | – |
agglo | 0.612 | 0.669 | 0.767 | 0.836 | 0.862 | 1.56 | 1.36 | 0.728 | 0.760 | 0.777 |
flat | 0.608 | 0.658 | 0.753 | 0.830 | 0.859 | 1.63 | 1.42 | 0.726 | 0.760 | 0.776 |
oriented mean [4] | 0.584 | 0.643 | 0.741 | 0.824 | 0.854 | 1.71 | 1.49 | 0.725 | 0.759 | 0.758 |
mean | 0.540 | 0.597 | 0.694 | 0.791 | 0.834 | 1.80 | 1.63 | 0.643 | 0.666 | 0.689 |
ODS uses the optimal scale for the entire dataset while OIS uses the optimal scale for each image.