Deploying accurate object detection for Vulnerable Road User (VRU) safety on edge hardware requires balancing model capacity against computational constraints. Large models achieve high accuracy but fail under INT8 quantization required for edge deployment, while small models sacrifice detection performance. This paper presents a knowledge distillation (KD) framework that trains a compact YOLOv8-S student (11.2M parameters) to mimic a YOLOv8-L teacher (43.7M parameters), achieving 3.9x compression while preserving quantization robustness. We evaluate on full-scale BDD100K (70K training images) with Post-Training Quantization to INT8. The teacher suffers catastrophic degradation under INT8 (-23% mAP), while the KD student retains accuracy (-5.6% mAP). Analysis reveals that KD transfers precision calibration rather than raw detection capacity: the KD student achieves 0.748 precision versus 0.653 for direct training at INT8, a 14.5% gain at equivalent recall, reducing false alarms by 44% versus the collapsed teacher. At INT8, the KD student exceeds the teacher's FP32 precision (0.748 vs. 0.718) in a model 3.9x smaller. These findings establish knowledge distillation as a requirement for deploying accurate, safety-critical VRU detection on edge hardware.

翻译：在边缘硬件上部署用于弱势道路使用者（VRU）安全检测的精确目标检测模型，需要在模型能力与计算约束之间取得平衡。大型模型虽能实现高精度，但在边缘部署所需的INT8量化下性能下降，而小型模型则牺牲检测性能。本文提出一种知识蒸馏（KD）框架，训练紧凑型YOLOv8-S学生模型（1120万参数）模仿YOLOv8-L教师模型（4370万参数），实现3.9倍压缩同时保持量化鲁棒性。我们在全量BDD100K数据集（7万张训练图像）上采用训练后量化（PTQ）至INT8进行评估。教师模型在INT8下出现严重性能退化（mAP下降23%），而KD学生模型保持精度（mAP下降5.6%）。分析表明，KD传递的是精度校准能力而非原始检测能力：KD学生在INT8下精确率达0.748，而直接训练模型仅为0.653（在同等召回率下提升14.5%），且相较于崩溃的教师模型，误报率降低44%。在INT8下，KD学生模型以3.9倍更小的模型规模，超越了教师模型FP32的精确率（0.748 vs. 0.718）。这些发现确立了知识蒸馏作为在边缘硬件上部署精确、安全关键的VRU检测模型的必要条件。