LiDAS：面向夜间感知的光照驱动动态主动传感
LiDAS: Lighting-driven Dynamic Active Sensing for Nighttime Perception
把车灯当作可主动控制的视觉执行器，用下游感知损失驱动光照场的预测，使白天训练的模型零样本迁移到夜间；这一「感知-照明闭环」思路可复用到其他主动传感（如可控曝光、主动红外补光）与感知-控制联合优化问题中。
原文依据 LiDAS enables zero-shot nighttime generalization of daytime-trained models through adaptive illumination control.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论夜间环境给基于相机的感知带来显著挑战，因为现有方法被动地依赖场景光照。我们提出光照驱动动态主动传感（LiDAS），一个将现成视觉感知模型与高清车灯相结合的闭环主动照明系统。LiDAS 并非均匀照亮场景，而是动态预测一个最优光照场以最大化下游感知性能，即在空旷区域降低光照，将其重新分配到目标区域。LiDAS 通过自适应光照控制，使白天训练的模型实现零样本夜间泛化。LiDAS 在合成数据上训练，并在真实世界闭环驾驶场景中零样本部署，在同等功率下相比标准近光灯取得 +18.7% mAP50 与 +5.0% 平均交并比的提升。它在保持性能的同时将能耗降低 40%。LiDAS 可与域泛化方法互补，无需重新训练即可进一步增强鲁棒性。通过将现成的车灯转变为主动视觉执行器，LiDAS 为鲁棒的夜间感知提供了一种高性价比的解决方案。项目主页：https://simondemoreau.github.io/LiDAS/
英文原文摘要 1196 字符
Nighttime environments pose significant challenges for camera-based perception, as existing methods passively rely on the scene lighting. We introduce Lighting-driven Dynamic Active Sensing (LiDAS), a closed-loop active illumination system that combines off-the-shelf visual perception models with high-definition headlights. Rather than uniformly brightening the scene, LiDAS dynamically predicts an optimal illumination field that maximizes downstream perception performance, i.e., decreasing light on empty areas to reallocate it on object regions. LiDAS enables zero-shot nighttime generalization of daytime-trained models through adaptive illumination control. Trained on synthetic data and deployed zero-shot in real-world closed-loop driving scenarios, LiDAS enables +18.7% mAP50 and +5.0% mIoU over standard low-beam at equal power. It maintains performances while reducing energy use by 40%. LiDAS complements domain-generalization methods, further strengthening robustness without retraining. By turning readily available headlights into active vision actuators, LiDAS offers a cost-effective solution to robust nighttime perception. Project page: https://simondemoreau.github.io/LiDAS/