Scale-Invariant Vote-Based 3D Recognition and Registration from Point Clouds

Pham, Minh-Tri; Woodford, Oliver J.; Perbet, Frank; Maki, Atsuto; Gherardi, Riccardo; Stenger, Björn; Cipolla, Roberto

doi:10.1007/978-3-642-28661-2_6

Cited by 1 publication

(1 citation statement)

References 43 publications

(103 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After trimming excess soft tissue areas, such as gingiva, the data were superimposed using the best-fit method (vote-based pose estimation). Vote-based pose estimation is an algorithm that is used for 3D data superimposition [ 22 ]. In this method, the polygons of the entire dataset with minimal deviations are used for superimposition.…”

Section: Methodsmentioning

confidence: 99%

The effect of scanning pathways on trueness and precision in full-arch optical impression

et al. 2023

View full text Add to dashboard Cite

Background In this study, we investigated the effects of differences in scanning pathways during optical impression on the trueness and precision of full-arch impressions. Methods Reference data were obtained using a laboratory scanner. All optical impressions were measured across the dental arch using TRIOS® 3 in four different pathways. The reference and optical impression data were superimposed using the best-fit method. The criteria for superimposition were based on the starting side of the dental arch (partial arch best-fit method, PB) and based on the full arch (full arch best-fit method, FB). The data were compared between the left and right molars (starting and ending sides). The scan deviations for trueness (n = 5) and precision (n = 10) were obtained for each group by calculating the root mean square (RMS) of the deviation at each measurement point. Visual observations using superimposed color map images revealed variations in trueness. Results There were no significant differences in scanning time or amount of scan data between the four scanning pathways. Trueness did not differ significantly among the four pathways with respect to the starting and ending sides, regardless of the superimposition criteria. Precision with PB was significantly different between scanning pathways A and B, and pathways B and C for the starting sides, and between scanning pathways A and B, and pathways A and D for the ending sides. In contrast, there was no significant difference between the starting and ending side in pathways for FB. Regarding PB, color map images showed a large error range in the direction toward the molar radius for the occlusal surface and cervical regions on the ending sides. Conclusion Differences in the scanning pathways did not affect trueness, regardless of the superimposition criteria. On the other hand, differences in the scanning pathways affected the precision of the starting and ending sides with PB. Scanning pathways B and D were more precise on the starting and ending sides, respectively.

show abstract

Section: Methodsmentioning

confidence: 99%