XD-RCDepth: Lightweight Radar-Camera Depth Estimation with Explainability-Aligned and Distribution-Aware Distillation
This work addresses depth estimation in autonomous driving by improving efficiency and interpretability, though it appears incremental as it builds on existing fusion and distillation methods.
The paper tackles depth estimation for autonomous driving by proposing XD-RCDepth, a lightweight radar-camera fusion model that reduces parameters by 29.7% compared to the state-of-the-art lightweight baseline while maintaining comparable accuracy, with a 7.97% reduction in MAE through knowledge-distillation strategies.
Depth estimation remains central to autonomous driving, and radar-camera fusion offers robustness in adverse conditions by providing complementary geometric cues. In this paper, we present XD-RCDepth, a lightweight architecture that reduces the parameters by 29.7% relative to the state-of-the-art lightweight baseline while maintaining comparable accuracy. To preserve performance under compression and enhance interpretability, we introduce two knowledge-distillation strategies: an explainability-aligned distillation that transfers the teacher's saliency structure to the student, and a depth-distribution distillation that recasts depth regression as soft classification over discretized bins. Together, these components reduce the MAE compared with direct training with 7.97% and deliver competitive accuracy with real-time efficiency on nuScenes and ZJU-4DRadarCam datasets.