Differentiable Stereopsis: Meshes from multiple views using differentiable rendering
This addresses the problem of 3D reconstruction from limited views for applications in computer vision and graphics, representing an incremental improvement by integrating existing techniques.
The paper tackles the problem of reconstructing 3D meshes from few input views and noisy cameras by proposing Differentiable Stereopsis, which combines traditional stereopsis with differentiable rendering to predict textured meshes for objects with varying topologies and shape, showing compelling reconstructions on challenging real-world scenes.
We propose Differentiable Stereopsis, a multi-view stereo approach that reconstructs shape and texture from few input views and noisy cameras. We pair traditional stereopsis and modern differentiable rendering to build an end-to-end model which predicts textured 3D meshes of objects with varying topologies and shape. We frame stereopsis as an optimization problem and simultaneously update shape and cameras via simple gradient descent. We run an extensive quantitative analysis and compare to traditional multi-view stereo techniques and state-of-the-art learning based methods. We show compelling reconstructions on challenging real-world scenes and for an abundance of object types with complex shape, topology and texture. Project webpage: https://shubham-goel.github.io/ds/