Parameterization-driven Neural Surface Reconstruction for Object-oriented Editing in Neural Rendering
This addresses the need for object-oriented editing in neural rendering, particularly for applications like human heads and man-made objects, though it appears incremental as it builds on existing neural rendering pipelines.
This paper tackles the problem of enabling intuitive editing of 3D objects in neural rendering by introducing a neural algorithm that parameterizes neural implicit surfaces to simple parametric domains like spheres and polycubes, allowing users to specify domain configurations that closely match object geometry and ensuring high-quality mapping with minimal distortion.
The advancements in neural rendering have increased the need for techniques that enable intuitive editing of 3D objects represented as neural implicit surfaces. This paper introduces a novel neural algorithm for parameterizing neural implicit surfaces to simple parametric domains like spheres and polycubes. Our method allows users to specify the number of cubes in the parametric domain, learning a configuration that closely resembles the target 3D object's geometry. It computes bi-directional deformation between the object and the domain using a forward mapping from the object's zero level set and an inverse deformation for backward mapping. We ensure nearly bijective mapping with a cycle loss and optimize deformation smoothness. The parameterization quality, assessed by angle and area distortions, is guaranteed using a Laplacian regularizer and an optimized learned parametric domain. Our framework integrates with existing neural rendering pipelines, using multi-view images of a single object or multiple objects of similar geometries to reconstruct 3D geometry and compute texture maps automatically, eliminating the need for any prior information. We demonstrate the method's effectiveness on images of human heads and man-made objects.