Explicit Neural Surfaces: Learning Continuous Geometry With Deformation Fields
This work addresses surface reconstruction for computer vision and graphics applications, offering an incremental improvement in efficiency over existing methods.
The paper tackles the problem of reconstructing explicit surfaces from multiple views by introducing Explicit Neural Surfaces (ENS), which uses neural deformation fields to transform a base domain into target shapes, resulting in faster training and inference times compared to implicit surfaces while maintaining competitive reconstruction performance.
We introduce Explicit Neural Surfaces (ENS), an efficient smooth surface representation that directly encodes topology with a deformation field from a known base domain. We apply this representation to reconstruct explicit surfaces from multiple views, where we use a series of neural deformation fields to progressively transform the base domain into a target shape. By using meshes as discrete surface proxies, we train the deformation fields through efficient differentiable rasterization. Using a fixed base domain allows us to have Laplace-Beltrami eigenfunctions as an intrinsic positional encoding alongside standard extrinsic Fourier features, with which our approach can capture fine surface details. Compared to implicit surfaces, ENS trains faster and has several orders of magnitude faster inference times. The explicit nature of our approach also allows higher-quality mesh extraction whilst maintaining competitive surface reconstruction performance and real-time capabilities.