LiDAS: Lighting-driven Dynamic Active Sensing for Nighttime Perception

2025-12-09

AI Summary

Nighttime visual perception is fundamentally constrained by passive illumination, limiting adaptability to downstream task requirements. This paper proposes a closed-loop dynamic active illumination system that synergistically coordinates a vehicle-mounted programmable headlamp with off-the-shelf vision models (e.g., YOLO, SegFormer) to generate, in real time, an optimal illumination field—focused on targets and suppressing irrelevant regions. We introduce the first illumination-driven zero-shot nighttime generalization mechanism, enabling seamless transfer of daytime-trained models to real-world nighttime scenes without model retraining. Illumination policy and perception performance are jointly optimized; the policy network is trained on synthetic data. Experiments demonstrate: (i) +18.7% mAP₅₀ and +5.0% mIoU over standard low-beam headlights at equal power; (ii) no performance degradation when power is reduced by 40%; and (iii) robust, plug-and-play zero-shot deployment on real vehicles.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Learning to Search
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

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.

Problem

• Optimizes illumination to enhance nighttime perception performance
• Enables zero-shot generalization of daytime models to nighttime
• Reduces energy consumption while maintaining perception accuracy

Innovation

• Dynamic optimal illumination field prediction
• Zero-shot nighttime generalization via adaptive lighting
• Energy-efficient active illumination using headlights