Table 6 Recognition rates and and per-object detection times (where available) for the Queen’s scenes
From: Local shape feature fusion for improved matching, pose estimation and 3D object recognition
Method | Recognition rate | Timing (s) |
|---|---|---|
External systems | ||
VD-LSD (SQ) | 0.838 | 2.964 |
EM | 0.824 | – |
RoPS based | 0.954 | 153.8 |
RANSAC, single feature | ||
ECSAD | 0.791 | 0.659 |
NDHist | 0.430 | 0.495 |
RoPS | 0.676 | 0.766 |
SHOT | 0.841 | 0.742 |
SI | 0.618 | 0.565 |
RANSAC, ternary feature | ||
ECSAD+NDHist+RoPS | 0.845 | 0.788 |
ECSAD+NDHist+SHOT | 0.826 | 0.719 |
ECSAD+NDHist+SI | 0.836 | 0.632 |
ECSAD+RoPS+SHOT | 0.894 | 0.753 |
ECSAD+RoPS+SI | 0.894 | 0.698 |
ECSAD+SHOT+SI | 0.913 | 0.679 |
NDHist+RoPS+SHOT | 0.836 | 0.674 |
NDHist+RoPS+SI | 0.802 | 0.627 |
NDHist+SHOT+SI | 0.826 | 0.626 |
RoPS+SHOT+SI | 0.879 | 0.797 |