Abstract
Semantic segmentation models are often affected by illumination changes, and fail to predict correct labels. Although there has been a lot of research on indoor semantic segmentation, it has not been studied in low-light environments. In this paper we propose a new framework, LISU, for Low-light Indoor Scene Understanding. We first decompose the low-light images into reflectance and illumination components, and then jointly learn reflectance restoration and semantic segmentation. To train and evaluate the proposed framework, we propose a new data set, namely LLRGBD, which consists of a large synthetic low-light indoor data set (LLRGBD-synthetic) and a small real data set (LLRGBD-real). The experimental results show that the illumination-invariant features effectively improve the performance of semantic segmentation. Compared with the baseline model, the mIoU of the proposed LISU framework has increased by 11.5%. In addition, pre-training on our synthetic data set increases the mIoU by 7.2%. Our data sets and models are available on our project website.
Original language | English |
---|---|
Title of host publication | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
Editors | N. Paparoditis, C. Mallet, F. Lafarge, M.Y. Yang, A. Yilmaz, J.D. Wegner, F. Remondino, T. Fuse, I. Toschi |
Publisher | International Society for Photogrammetry and Remote Sensing (ISPRS) |
Pages | 427-432 |
Number of pages | 6 |
Volume | XLIII-B2-2021 |
Edition | B2-2021 |
DOIs | |
Publication status | Published - 28 Jun 2021 |
Event | 24th ISPRS Congress "Imaging Today, Foreseeing Tomorrow", Commission I 2021 - Virtual Event, Nice Virtual, France Duration: 5 Jul 2021 → 9 Jul 2021 Conference number: 24 https://www.isprs2020-nice.com/ |
Conference
Conference | 24th ISPRS Congress "Imaging Today, Foreseeing Tomorrow", Commission I 2021 |
---|---|
Country/Territory | France |
City | Nice Virtual |
Period | 5/07/21 → 9/07/21 |
Internet address |