Geometry-Consistent Neural Shape Representation with Implicit Displacement Fields
arXiv:2106.05187
Abstract
We present implicit displacement fields, a novel representation for detailed 3D geometry. Inspired by a classic surface deformation technique, displacement mapping, our method represents a complex surface as a smooth base surface plus a displacement along the base's normal directions, resulting in a frequency-based shape decomposition, where the high frequency signal is constrained geometrically by the low frequency signal. Importantly, this disentanglement is unsupervised thanks to a tailored architectural design that has an innate frequency hierarchy by construction. We explore implicit displacement field surface reconstruction and detail transfer and demonstrate superior representational power, training stability and generalizability.
Accepted to ICLR 2022, 20 pages including appendix. Code available at https://github.com/yifita/idf
Cited by in corpus (5)
- Physics Informed Neural Fields for Smoke Reconstruction with Sparse Data
- Robust Zero Level-Set Extraction from Unsigned Distance Fields Based on Double Covering
- Geometric implicit neural representations for signed distance functions
- Spectral Total-Variation Processing of Shapes: Theory and Applications
- Theoretical and Empirical Analysis of a Fast Algorithm for Extracting Polygons from Signed Distance Bounds