Geometry-Consistent Neural Shape Representation with Implicit Displacement Fields

ICLR 2022
Wang Yifan, , Olga Sorkine-Hornung,

We represent detailed geometries as a sum of a coarse base shape represented as low-frequency signed distance function and a high-frequency implicit displacement field, which offsets the base iso-surface along the base’s normal directions.


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.


accompanying video


This work was supported in part by Apple's AI/ML PhD fellowship program.