Shape from Projections via Differentiable Forward Projector for Computed Tomography
This addresses shape reconstruction in computed tomography, offering a novel approach for improved accuracy in medical or material science imaging, though it appears incremental as it builds on differentiable rendering techniques.
The paper tackles 3D shape reconstruction from projections in computed tomography by proposing a mesh-based method with a differentiable forward projector, showing it outperforms traditional voxel-based methods on noisy simulated data and applies to real electron tomography images.
In computed tomography, the reconstruction is typically obtained on a voxel grid. In this work, however, we propose a mesh-based reconstruction method. For tomographic problems, 3D meshes have mostly been studied to simulate data acquisition, but not for reconstruction, for which a 3D mesh means the inverse process of estimating shapes from projections. In this paper, we propose a differentiable forward model for 3D meshes that bridge the gap between the forward model for 3D surfaces and optimization. We view the forward projection as a rendering process, and make it differentiable by extending recent work in differentiable rendering. We use the proposed forward model to reconstruct 3D shapes directly from projections. Experimental results for single-object problems show that the proposed method outperforms traditional voxel-based methods on noisy simulated data. We also apply the proposed method on electron tomography images of nanoparticles to demonstrate the applicability of the method on real data.