Towards Learning Low-Light Indoor Semantic Segmentation with Illumination-Invariant Features

N. Zhang*, F.C. Nex, N. Kerle, G. Vosselman

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

4 Citations (Scopus)
81 Downloads (Pure)


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 languageEnglish
Title of host publicationThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
EditorsN. Paparoditis, C. Mallet, F. Lafarge, M.Y. Yang, A. Yilmaz, J.D. Wegner, F. Remondino, T. Fuse, I. Toschi
PublisherInternational Society for Photogrammetry and Remote Sensing (ISPRS)
Number of pages6
Publication statusPublished - 28 Jun 2021
Event24th ISPRS Congress "Imaging Today, Foreseeing Tomorrow", Commission I 2021 - Virtual Event, Nice Virtual, France
Duration: 5 Jul 20219 Jul 2021
Conference number: 24


Conference24th ISPRS Congress "Imaging Today, Foreseeing Tomorrow", Commission I 2021
CityNice Virtual
Internet address


Dive into the research topics of 'Towards Learning Low-Light Indoor Semantic Segmentation with Illumination-Invariant Features'. Together they form a unique fingerprint.

Cite this