Indoor 3D scene reconstruction using plane-regularized signed distance field with P2SDF
Indoor 3D scene reconstruction using plane-regularized signed distance field with P2SDF
P2SDF for Neural Indoor Scene Reconstruction
arXiv paper abstract https://arxiv.org/abs/2303.00236
arXiv PDF paper https://arxiv.org/pdf/2303.00236.pdf
Given only a set of images, neural implicit surface representation ... in 3D surface reconstruction ... usually fail in low-textured regions, including the floors, walls
... low-textured regions usually correspond to planes ... propose to leverage a novel Pseudo Plane-regularized Signed Distance Field (P2SDF) for indoor scene reconstruction ... consider adjacent pixels with similar colors to be on the same pseudo planes.
... signed distances of the points on the planes are regularized by the estimated plane parameters in the training phase.
... unsupervised plane segments are ... noisy ... propose to assign different weights to the sampled points on the plane in plane estimation as well as the regularization loss.
... rays in the planar regions are redundant ... propose a keypoint-guided rays sampling strategy that attends to the informative textured regions with large color variations, and the implicit network gets a better reconstruction, compared with the original uniform ray sampling strategy.
... P2SDF achieves competitive reconstruction performance in Manhattan scenes ...
Please like and share this post if you enjoyed it using the buttons at the bottom!
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
留言