Nighttime camera-based depth estimation is a highly challenging task, especially for autonomous driving applications, where accurate depth perception is essential for ensuring safe navigation. Models trained on daytime data often fail in the absence of precise but costly LiDAR. Even vision foundation models trained on large amounts of data are unreliable in low-light conditions. In this work, we aim to improve the reliability of perception systems at night time. To this end, we introduce Light Enhanced Depth (LED), a novel, cost-effective approach that significantly improves depth estimation in low-light environments by harnessing a pattern projected by high definition headlights available in modern vehicles. LED leads to significant performance boosts across multiple depth-estimation architectures (encoder-decoder, Adabins, DepthFormer, Depth Anything V2) both on synthetic and real datasets. Furthermore, increased performances beyond illuminated areas reveal a holistic enhancement in scene understanding. Finally, we release the Nighttime Synthetic Drive Dataset, a synthetic and photo-realistic nighttime dataset, which comprises 49,990 comprehensively annotated images.

翻译：基于摄像头的夜间深度估计是一项极具挑战性的任务，尤其在自动驾驶应用中，精确的深度感知对于确保安全导航至关重要。在缺乏精确但昂贵的激光雷达的情况下，基于白天数据训练的模型往往失效。即使在大量数据上训练的视觉基础模型，在低光照条件下也表现不可靠。本研究旨在提升夜间感知系统的可靠性。为此，我们提出了光增强深度估计（LED），一种新颖且成本效益高的方法，通过利用现代车辆配备的高清前照灯投射的图案，显著改善了低光照环境下的深度估计性能。LED在多种深度估计架构（编码器-解码器、Adabins、DepthFormer、Depth Anything V2）上，无论是合成数据集还是真实数据集，均带来了显著的性能提升。此外，在照明区域之外的性能增强揭示了场景理解的整体改善。最后，我们发布了夜间合成驾驶数据集，这是一个包含49,990张全面标注图像的合成且具有照片级真实感的夜间数据集。