MOSE: NeRFによるノイズを含む単眼事前情報を持ち上げたセマンティック再構成
MOSE: Monocular Semantic Reconstruction Using NeRF-Lifted Noisy Priors
単眼画像からノイズの多い2Dセマンティック事前情報をNeRFで3Dに持ち上げ、クラス非依存のセグメンテーションマスクをガイドに使うことで、3D/2Dのセマンティックセグメンテーションと表面再構成を同時に高精度化する手法を提案。
著者: Zhenhua Du, Binbin Xu, Haoyu Zhang, Kai Huo, Shuaifeng Zhi
分類: cs.CV, cs.AI, cs.RO
原文アブストラクト
Accurately reconstructing dense and semantically annotated 3D meshes from monocular images remains a challenging task due to the lack of geometry guidance and imperfect view-dependent 2D priors. Though we have witnessed recent advancements in implicit neural scene representations enabling precise 2D rendering simply from multi-view images, there have been few works addressing 3D scene understanding with monocular priors alone. In this paper, we propose MOSE, a neural field semantic reconstruction approach to lift inferred image-level noisy priors to 3D, producing accurate semantics and geometry in both 3D and 2D space. The key motivation for our method is to leverage generic class-agnostic segment masks as guidance to promote local consistency of rendered semantics during training. With the help of semantics, we further apply a smoothness regularization to texture-less regions for better geometric quality, thus achieving mutual benefits of geometry and semantics. Experiments on the ScanNet dataset show that our MOSE outperforms relevant baselines across all metrics on tasks of 3D semantic segmentation, 2D semantic segmentation and 3D surface reconstruction.
