Tutorial demo of the new plane constraint functionality in Phi.3D 2.2 on the DPI-8X and DPI-8X SR handheld 3D scanners.
More Information:
Parallel Plane Constraints are a useful tool to limit drift in large captures. Although any capture can benefit they are particularly useful when scanning large stretches without loop closure. The operator may pick various planes at scanning time. During Global Optimization the picked planes are then clustered by their orientation and the scene is optimized in a manner that makes all planes within a cluster perfectly parallel. Warning: Some care has to be taken when using this feature: If the scanner operator picks surfaces to be parallel which aren't in reality and then runs Global Optimization using Parallel Plane Constraints, loss of accuracy and/or map corruption may be the result.
In Summary:
Use the Parallel Plane Constraints feature when you think that the drift in the capture exceeds any non-parallelism (of constrained planes) in the real world and using surveyed tags/targets is not an option. For example: when capturing multi-level staircases there may be some degrees of angular discrepancy between the first and the last landing of the staircase. Parallel Plane Constraints can be used to make these landings parallel again during Global Optimization, by picking the landing planes during capture.
More Information:
Parallel Plane Constraints are a useful tool to limit drift in large captures. Although any capture can benefit they are particularly useful when scanning large stretches without loop closure. The operator may pick various planes at scanning time. During Global Optimization the picked planes are then clustered by their orientation and the scene is optimized in a manner that makes all planes within a cluster perfectly parallel. Warning: Some care has to be taken when using this feature: If the scanner operator picks surfaces to be parallel which aren't in reality and then runs Global Optimization using Parallel Plane Constraints, loss of accuracy and/or map corruption may be the result.
In Summary:
Use the Parallel Plane Constraints feature when you think that the drift in the capture exceeds any non-parallelism (of constrained planes) in the real world and using surveyed tags/targets is not an option. For example: when capturing multi-level staircases there may be some degrees of angular discrepancy between the first and the last landing of the staircase. Parallel Plane Constraints can be used to make these landings parallel again during Global Optimization, by picking the landing planes during capture.