SparseNeuS: Fast Generalizable Neural Surface Reconstruction from Sparse Views
This addresses the challenge of incomplete or distorted reconstructions in sparse-view scenarios for applications like 3D modeling, with incremental improvements in generalization and efficiency.
The paper tackles the problem of surface reconstruction from sparse multi-view images, where existing methods often fail, by introducing SparseNeuS, which achieves high-quality reconstruction with as few as 2-3 images and outperforms state-of-the-art methods in experiments.
We introduce SparseNeuS, a novel neural rendering based method for the task of surface reconstruction from multi-view images. This task becomes more difficult when only sparse images are provided as input, a scenario where existing neural reconstruction approaches usually produce incomplete or distorted results. Moreover, their inability of generalizing to unseen new scenes impedes their application in practice. Contrarily, SparseNeuS can generalize to new scenes and work well with sparse images (as few as 2 or 3). SparseNeuS adopts signed distance function (SDF) as the surface representation, and learns generalizable priors from image features by introducing geometry encoding volumes for generic surface prediction. Moreover, several strategies are introduced to effectively leverage sparse views for high-quality reconstruction, including 1) a multi-level geometry reasoning framework to recover the surfaces in a coarse-to-fine manner; 2) a multi-scale color blending scheme for more reliable color prediction; 3) a consistency-aware fine-tuning scheme to control the inconsistent regions caused by occlusion and noise. Extensive experiments demonstrate that our approach not only outperforms the state-of-the-art methods, but also exhibits good efficiency, generalizability, and flexibility.