Differentiable Rendering of Neural SDFs through Reparameterization
This addresses a bottleneck in 3D reconstruction and inverse rendering for computer vision and graphics researchers, offering an incremental improvement by adapting area-sampling techniques to neural SDFs.
The paper tackles the problem of computing correct gradients for geometric scene parameters in neural SDF renderers, which is challenging due to discontinuities at object silhouettes. The result is a differentiable renderer that enables optimization of neural shapes from multi-view images, producing comparable 3D reconstructions to recent methods without needing 2D segmentation masks or volumetric approximations.
We present a method to automatically compute correct gradients with respect to geometric scene parameters in neural SDF renderers. Recent physically-based differentiable rendering techniques for meshes have used edge-sampling to handle discontinuities, particularly at object silhouettes, but SDFs do not have a simple parametric form amenable to sampling. Instead, our approach builds on area-sampling techniques and develops a continuous warping function for SDFs to account for these discontinuities. Our method leverages the distance to surface encoded in an SDF and uses quadrature on sphere tracer points to compute this warping function. We further show that this can be done by subsampling the points to make the method tractable for neural SDFs. Our differentiable renderer can be used to optimize neural shapes from multi-view images and produces comparable 3D reconstructions to recent SDF-based inverse rendering methods, without the need for 2D segmentation masks to guide the geometry optimization and no volumetric approximations to the geometry.