Calibrated and Resource-Aware Super-Resolution for Reliable Driver Behavior Analysis
This addresses safety-critical reliability issues in driver behavior analysis, representing an incremental improvement over existing low-resolution training methods.
The paper tackles the problem of unreliable confidence scores in driver monitoring systems by proposing a resource-aware adaptive super-resolution framework that optimizes for model calibration and precision-recall on critical events. It achieves state-of-the-art performance with a calibration ECE of 5.8% (vs. 6.2% for baselines), AUPR of 0.78 for drowsiness detection (vs. 0.74), and precision-recall of 0.74 for phone use detection (vs. 0.71).
Driver monitoring systems require not just high accuracy but reliable, well-calibrated confidence scores for safety-critical deployment. While direct low-resolution training yields high overall accuracy, it produces poorly calibrated predictions that can be dangerous in safety-critical scenarios. We propose a resource-aware adaptive super-resolution framework that optimizes for model calibration and high precision-recall on critical events. Our approach achieves state-of-the-art performance on safety-centric metrics: best calibration (ECE of 5.8\% vs 6.2\% for LR-trained baselines), highest AUPR for drowsiness detection (0.78 vs 0.74), and superior precision-recall for phone use detection (0.74 vs 0.71). A lightweight artifact detector (0.3M parameters, 5.2ms overhead) provides additional safety by filtering SR-induced hallucinations. While LR-trained video models serve as strong general-purpose baselines, our adaptive framework represents the state-of-the-art solution for safety-critical applications where reliability is paramount.