PVP-Recon: Progressive View Planning via Warping Consistency for Sparse-View Surface Reconstruction

• DOI: 10.1145/3687896
• Publication Date: 2024-11-19T15:46:04Z
• Authors: Sheng Ye; Yuze He; Matthieu Lin; Jenny Sheng; Ruoyu Fan; Yiheng Han; Yubin Hu; Ran Yi; Yu-Hui Wen; Yong-Jin Liu; Wenping Wang
• Keywords: Image warping; View synthesis

Neural implicit representations have revolutionized dense multi-view surface reconstruction, yet their performance significantly diminishes with sparse input views. A few pioneering works sought to tackle this challenge by leveraging additional geometric priors or multi-scene generalizability. However, they are still hindered the imperfect choice of views, using images under empirically determined viewpoints. We propose PVP-Recon , a novel and effective sparse-view reconstruction method that progressively plans next best views form an optimal set viewpoints for image capturing. starts initial as 3 adds new which based on warping score reflects information gain each newly added view. This progressive view planning progress is interleaved neural SDF-based module utilizes multi-resolution hash features, enhanced training scheme directional Hessian loss. Quantitative qualitative experiments three benchmark datasets show our system achieves high-quality constrained budget outperforms existing baselines.