Neural Representation of Open Surfaces

Christiansen, Thor V.; Bærentzen, Jakob Andreas; Paulsen, Rasmus R.; Hannemose, Morten R.

Neural Representation of Open Surfaces

dc.contributor.author	Christiansen, Thor V.	en_US
dc.contributor.author	Bærentzen, Jakob Andreas	en_US
dc.contributor.author	Paulsen, Rasmus R.	en_US
dc.contributor.author	Hannemose, Morten R.	en_US
dc.contributor.editor	Memari, Pooran	en_US
dc.contributor.editor	Solomon, Justin	en_US
dc.date.accessioned	2023-06-30T06:19:17Z
dc.date.available	2023-06-30T06:19:17Z
dc.date.issued	2023
dc.description.abstract	Neural implicit surfaces have emerged as an effective, learnable representation for shapes of arbitrary topology. However, representing open surfaces remains a challenge. Different methods, such as unsigned distance fields (UDF), have been proposed to tackle this issue, but a general solution remains elusive. The generalized winding number (GWN), which is often used to distinguish interior points from exterior points of 3D shapes, is arguably the most promising approach. The GWN changes smoothly in regions where there is a hole in the surface, but it is discontinuous at points on the surface. Effectively, this means that it can be used in lieu of an implicit surface representation while providing information about holes, but, unfortunately, it does not provide information about the distance to the surface necessary for e.g. ray tracing, and special care must be taken when implementing surface reconstruction. Therefore, we introduce the semi-signed distance field (SSDF) representation which comprises both the GWN and the surface distance. We compare the GWN and SSDF representations for the applications of surface reconstruction, reconstruction from partial data, interpolation, and latent vector analysis using two very different data sets. We find that both the GWN and SSDF are well suited for neural representation of open surfaces.	en_US
dc.description.number	5
dc.description.sectionheaders	Representation and Learning
dc.description.seriesinformation	Computer Graphics Forum
dc.description.volume	42
dc.identifier.doi	10.1111/cgf.14916
dc.identifier.issn	1467-8659
dc.identifier.pages	13 pages
dc.identifier.uri	https://doi.org/10.1111/cgf.14916
dc.identifier.uri	https://diglib.eg.org:443/handle/10.1111/cgf14916
dc.publisher	The Eurographics Association and John Wiley & Sons Ltd.	en_US
dc.subject	CCS Concepts: Neural Implicit Surface Representations of open surfaces -> Generalized Winding Number Field, Semi-Signed Distance Fields
dc.subject	Neural Implicit Surface Representations of open surfaces
dc.subject	Generalized Winding Number Field
dc.subject	Semi
dc.subject	Signed Distance Fields
dc.title	Neural Representation of Open Surfaces	en_US