Get 3D object shape with few views by learning surface geometry with NeuSurf
Get 3D object shape with few views by learning surface geometry with NeuSurf
NeuSurf: On-Surface Priors for Neural Surface Reconstruction from Sparse Input Views
arXiv paper abstract https://arxiv.org/abs/2312.13977
arXiv PDF paper https://arxiv.org/pdf/2312.13977.pdf
… neural implicit functions … demonstrated remarkable results in … multi-view reconstruction … However, most … methods are … for dense views and exhibit unsatisfactory performance … with sparse views.
… propose a novel sparse view reconstruction framework that leverages on-surface priors to achieve highly faithful surface reconstruction.
… design several constraints on global geometry alignment and local geometry refinement for jointly optimizing coarse shapes and fine details.
… train a neural network to learn a global implicit field from the on-surface points obtained from SfM and then leverage it as a coarse geometric constraint.
To exploit local geometric consistency, … project on-surface points onto seen and unseen views, treating the consistent loss of projected features as a fine geometric constraint.
… demonstrate significant improvements over the state-of-the-art methods.
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
LinkedIn https://www.linkedin.com/in/morris-lee-47877b7b
