a) input line drawing and mask b) stroke-aligned parametrization c) output curve network a) input line drawing and mask b) s a) input line drawing and mask a) input line drawing and mask ask b) stroke-aligned parametrization e drawing and mask b) stroke-aligned par and mask b) stroke-aligned parametrizati igned parametrization c) output curve n b) stroke-aligned parametrization roke-aligned parametrization c) output Figure 1: Starting from an input line drawing (left), we locally parametrize the sketch as a grid aligned with the strokes (middle). Neighboring parallel strokes are automatically snapped to the same isoline of the parametrization, while junctions are snapped to grid nodes. This parametrization facilitates the extraction of a clean network of Bézier curves (right). Using a simple mask, the user can locally specify the desired amount of simplification in the output (purple scribbles: less simplification, orange scribbles: more simplification). See supplemental materials for a result without the mask.

Figure 1: Freehand 3D sketching in AR/VR allows rapid conceptualization of design ideas (a). However, 3D inputs are prone to large inaccuracies (a, inset) and sketches cannot be utilized in downstream design pipelines. Our novel 3D sketching system, CASSIE, allows the creation of clean, well-connected 3D curve networks by performing automatic stroke neatening (b). These curve networks are augmented by our on-the-fly cycle detection and surfacing method (c) which improves shape perception by providing occlusion cues. We evaluated CASSIE with 12 users and utilized it for creating 3D concepts for a variety of application domains (d).

Recent surface acquisition technologies based on microsensors produce three-space tangential curve data which can be transformed into a network of space curves with surface normals. This paper addresses the problem of surfacing an arbitrary closed 3D curve network with given surface normals. Thanks to the normal vector input, the patch finding problem can be solved unambiguously and an initial piecewise smooth triangle mesh is computed. The input normals are propagated throughout the mesh. Together with the initial mesh, the propagated normals are used to compute mean curvature vectors. We then compute the final mesh as the solution of a new variational optimization method based on the mean curvature vectors. The intuition behind this original approach is to guide the standard Laplacian-based variational methods by the curvature information extracted from the input normals. The normal input increases shape fidelity and allows to achieve globally smooth and visually pleasing shapes.