SAID-NeRF: Segmentation-AIDed NeRF for Depth Completion of Transparent Objects

arXiv:2403.19607v1 Announce Type: cross
Abstract: Acquiring accurate depth information of transparent objects using off-the-shelf RGB-D cameras is a well-known challenge in Computer Vision and Robotics. Depth estimation/completion methods are typically employed and trained on datasets with quality depth labels acquired from either simulation, additional sensors or specialized data collection setups and known 3d models. However, acquiring reliable depth information for datasets at scale is not straightforward, limiting training scalability and generalization. Neural Radiance Fields (NeRFs) are learning-free approaches and have …

abstract acquired arxiv cameras challenge collection computer computer vision cs.cv cs.ro data data collection datasets information labels nerf objects quality rgb-d robotics segmentation sensors simulation transparent type vision