Urban Night Scenery Reconstruction by Day-night Registration and Synthesis

Abstract

Although large-scale 3D reconstruction by photogrammetry has been well studied and applied, the reconstruction of night scenery in urban areas has not been thoroughly considered. At night, low-light conditions often cause the images to lack sharpness and high-dynamic range issue leads to saturation. The SFM reconstruction pipeline that works well in daylight is likely to recover only limited dense points of bright fragmented objects near artificial lighting. Here, we propose a novel solution based on registration and synthesis between the night-time reconstruction and that of the same region in daytime. A registration pipeline is developed for conformal matching of the day and night point clouds. For the coarse registration step, we use detected plane features to search and match 4-plane congruent sets. For the fine registration step, we consider the positions of windows, a commonly-occurring object cue in urban building scenes as markers for accurate positioning. This leads to final registration error less than 0.2 degrees in rotation, and 0.2% in scale and translation. Finally, we synthesize the daytime textured model and the night point clouds to produce vivid visual effects of urban night scenery.