Get 3D scene from depth images using uncertainty and curvature to guide sampling and with Sang
Get 3D scene from depth images using uncertainty and curvature to guide sampling and with Sang
NeuroSURF: Neural Uncertainty-aware Robust Surface Reconstruction
arXiv paper abstract https://arxiv.org/abs/2306.02099
arXiv PDF paper https://arxiv.org/pdf/2306.02099.pdf
Neural implicit functions … popular for representing surfaces because … offer an adaptive resolution and support arbitrary topologies … previous works … often ignore the effect of input quality and sampling methods
… introduce NeuroSURF, which generates significantly improved qualitative and quantitative reconstructions driven by a novel sampling and interpolation technique.
… show that employing a sampling technique that considers the geometric characteristics of inputs can enhance the training process.
… introduce a strategy that efficiently computes differentiable geometric features, namely, mean curvatures, to augment the sampling phase during the training period.
… augment the neural implicit surface representation with uncertainty, which offers insights into the occupancy and reliability of the output signed distance value, thereby expanding representation capabilities into open surfaces.
… demonstrate that NeuroSURF leads to state-of-the-art reconstructions on both synthetic and real-world data.
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
