Get 3D scene from 2D images using signed distance and geometrical constraints with Guo
Get 3D scene from 2D images using signed distance and geometrical constraints with Guo
Neural 3D Scene Reconstruction from Multi-view Images without 3D Supervision
arXiv paper abstract https://arxiv.org/abs/2306.17643
arXiv PDF paper https://arxiv.org/pdf/2306.17643.pdf
Neural scene reconstruction methods have achieved impressive performance in reconstructing complex geometry and low-textured regions in large scenes.
… propose … reconstructs scenes without 3D supervision … perform differentiable volume rendering … using … 2D images as supervision.
… impose geometry to improve … reconstruction … impose plane constraints to improve the reconstruction quality of low-textured regions
… introduce a signed distance function (SDF) field, a color field, and a probability field to represent the scene, and optimize the fields under the differentiable ray marching to reconstruct the scene.
… impose geometric constraints that project 3D points on the surface to similar-looking regions … in different views … make large planes keep parallel or vertical to the wall or floor.
… achieves competitive reconstruction compared with … methods that use 3D information as supervision on the ScanNet dataset.
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
