Improving neural implicit surfaces geometry with patch warping
This addresses the challenge of learning high-frequency textures in neural rendering for 3D reconstruction, offering a significant but incremental improvement over existing techniques.
The paper tackled the problem of limited accuracy in neural implicit surfaces for multi-view 3D reconstruction by adding a direct photo-consistency term with patch warping, resulting in over 20% improvement on DTU and EPFL benchmarks compared to state-of-the-art unsupervised methods.
Neural implicit surfaces have become an important technique for multi-view 3D reconstruction but their accuracy remains limited. In this paper, we argue that this comes from the difficulty to learn and render high frequency textures with neural networks. We thus propose to add to the standard neural rendering optimization a direct photo-consistency term across the different views. Intuitively, we optimize the implicit geometry so that it warps views on each other in a consistent way. We demonstrate that two elements are key to the success of such an approach: (i) warping entire patches, using the predicted occupancy and normals of the 3D points along each ray, and measuring their similarity with a robust structural similarity (SSIM); (ii) handling visibility and occlusion in such a way that incorrect warps are not given too much importance while encouraging a reconstruction as complete as possible. We evaluate our approach, dubbed NeuralWarp, on the standard DTU and EPFL benchmarks and show it outperforms state of the art unsupervised implicit surfaces reconstructions by over 20% on both datasets.