Incremental Instance-Oriented 3D Semantic Mapping via RGB-D Cameras for Unknown Indoor Scene

Table 1

Comparison of run-time performance. FQ denotes the frequency recognition of when the input frame is performed, and the class probabilities of the 3D map are updated.

Method	Representation	FQ	FPS
SemanticFusion [11]	Dense	Every 10 frames	Under 8 Hz
Hermans et al. [34]	Dense	Every 6 frames	3 Hz
PanopicFusion [44]	Dense	Every 10 frames	4.3 Hz
Voxblox++ [16]	Instance-oriented	Every frame	1 Hz
Pham et al. [45]	Instance-oriented	Every frame	1 Hz
Fusion++ [29]	Instance-oriented	Every frame	4 Hz
Ours	Instance-oriented	Every frame	10.8 Hz