Fast Low-light Enhancement and Deblurring for 3D Dark Scenes
This work provides a significantly faster and more effective solution for reconstructing 3D scenes from degraded imagery, benefiting applications in robotics and augmented reality where real-time performance is crucial.
This paper addresses the problem of novel view synthesis from 3D scenes captured under low-light, noisy, and motion-blurred conditions. The proposed FLED-GS framework achieves 21x faster training and 11x faster rendering compared to state-of-the-art LuSh-NeRF.
Novel view synthesis from low-light, noisy, and motion-blurred imagery remains a valuable and challenging task. Current volumetric rendering methods struggle with compound degradation, and sequential 2D preprocessing introduces artifacts due to interdependencies. In this work, we introduce FLED-GS, a fast low-light enhancement and deblurring framework that reformulates 3D scene restoration as an alternating cycle of enhancement and reconstruction. Specifically, FLED-GS inserts several intermediate brightness anchors to enable progressive recovery, preventing noise blow-up from harming deblurring or geometry. Each iteration sharpens inputs with an off-the-shelf 2D deblurrer and then performs noise-aware 3DGS reconstruction that estimates and suppresses noise while producing clean priors for the next level. Experiments show FLED-GS outperforms state-of-the-art LuSh-NeRF, achieving 21$\times$ faster training and 11$\times$ faster rendering.