Online 3D Scene Reconstruction Using Neural Object Priors
It addresses the problem of efficient and complete online 3D object reconstruction for applications like robotics or AR, but is incremental as it builds on existing neural implicit representations.
This paper tackles online 3D scene reconstruction at the object level from RGB-D video by proposing a feature grid interpolation mechanism and an object library with shape priors, resulting in improved reconstruction accuracy and completeness compared to state-of-the-art neural implicit models.
This paper addresses the problem of reconstructing a scene online at the level of objects given an RGB-D video sequence. While current object-aware neural implicit representations hold promise, they are limited in online reconstruction efficiency and shape completion. Our main contributions to alleviate the above limitations are twofold. First, we propose a feature grid interpolation mechanism to continuously update grid-based object-centric neural implicit representations as new object parts are revealed. Second, we construct an object library with previously mapped objects in advance and leverage the corresponding shape priors to initialize geometric object models in new videos, subsequently completing them with novel views as well as synthesized past views to avoid losing original object details. Extensive experiments on synthetic environments from the Replica dataset, real-world ScanNet sequences and videos captured in our laboratory demonstrate that our approach outperforms state-of-the-art neural implicit models for this task in terms of reconstruction accuracy and completeness.