Computer Graphics and Visualization › Publications › Adaptive pointcloud segmentation for assisted interactions

In Proceedings of I3D

Adaptive pointcloud segmentation for assisted interactions

Harald Steinlechner, Bernhard Rainer, Michael Schwärzler, and Georg Haaser

Our novel interactive approach for shape detection in point clouds allows for sophisticated interactions: Left: A Lasso selection selects only points that lie on the support shape as shown in the top image. Points in front and back of the support shape are not selected (bottom). Middle: A volumetric brush selection is performed on the selected support shape (top). Points are only selected if they belong to the support shape and intersect the brush (bottom). Right: Interactive LoD increment interaction along the selected support shape (drawn in red). The top image shows the original rendering model of the point cloud; the bottom image shows the point cloud with the additional points.

Publisher Paper Cite More

In this work, we propose an interaction-driven approach streamlined to support and improve a wide range of real-time 2D interaction metaphors for arbitrarily large pointclouds based on detected primitive shapes. Rather than performing shape detection as a costly pre-processing step on the entire point cloud at once, a usercontrolled interaction determines the region that is to be segmented next. By keeping the size of the region and the number of points small, the algorithm produces meaningful results and therefore feedback on the local geometry within a fraction of a second. We can apply these finding for improved picking and selection metaphors in large point clouds, and propose further novel shape-assisted interactions that utilize this local semantic information to improve the user’s workflow.

More Information

Publisher Paper preprint

Citation

Harald Steinlechner, Bernhard Rainer, Michael Schwärzler, and Georg Haaser, Adaptive pointcloud segmentation for assisted interactions, In Proceedings of I3D, pp. 14:1–14:9, 2019.

BibTex 

@inproceedings{bib:steinlechner:2019,
    author       = { Steinlechner, Harald and Rainer, Bernhard and Schwärzler, Michael and Haaser, Georg },    
    title        = { Adaptive pointcloud segmentation for assisted interactions },
    booktitle    = { In Proceedings of I3D },
    year         = { 2019 },
    pages        = { 14:1--14:9 },
    doi          = { 10.1145/3306131.3317023 },
    dblp         = { conf/si3d/SteinlechnerRSH19 },
    url          = { https://publications.graphics.tudelft.nl/papers/104 },
}