Recently, there has been an increased interest in NeRF methods which reconstruct differentiable representation of three-dimensional scenes. One of the main limitations of such methods is their inability to assess the confidence of the model in its predictions. In this paper, we propose a new neural network model for the formation of extended vector representations, called uSF, which allows the model to predict not only color and semantic label of each point, but also estimate the corresponding values of uncertainty. We show that with a small number of images available for training, a model quantifying uncertainty performs better than a model without such functionality. Code of the uSF approach is publicly available at https://github.com/sevashasla/usf/.
@article{arxiv.2312.08012,
title = {uSF: Learning Neural Semantic Field with Uncertainty},
author = {Vsevolod Skorokhodov and Darya Drozdova and Dmitry Yudin},
journal= {arXiv preprint arXiv:2312.08012},
year = {2024}
}
Comments
12 pages, 4 figures. This a preprint of the Work accepted for publication in Optical Memory and Neural Networks (Information Optics), \copyright, copyright 2024, Optical Memory and Neural Networks; https://www.pleiades.online/en/journal/optmem/