SurfR: Surface Reconstruction with Multi-scale Attention
This work addresses the need for efficient and detailed surface reconstruction in 3D modeling and computer vision, though it appears incremental as it builds on existing implicit methods with specific improvements.
The authors tackled the problem of fast and accurate surface reconstruction from unorganized point clouds by proposing a new implicit representation that achieves the best accuracy-speed trade-off, being faster than all baselines with only a marginal loss in performance compared to the state-of-the-art.
We propose a fast and accurate surface reconstruction algorithm for unorganized point clouds using an implicit representation. Recent learning methods are either single-object representations with small neural models that allow for high surface details but require per-object training or generalized representations that require larger models and generalize to newer shapes but lack details, and inference is slow. We propose a new implicit representation for general 3D shapes that is faster than all the baselines at their optimum resolution, with only a marginal loss in performance compared to the state-of-the-art. We achieve the best accuracy-speed trade-off using three key contributions. Many implicit methods extract features from the point cloud to classify whether a query point is inside or outside the object. First, to speed up the reconstruction, we show that this feature extraction does not need to use the query point at an early stage (lazy query). Second, we use a parallel multi-scale grid representation to develop robust features for different noise levels and input resolutions. Finally, we show that attention across scales can provide improved reconstruction results.