通过剪枝与知识蒸馏压缩自动驾驶多任务模型，兼顾精度与实时性。
Compressing Multi-Task Model for Autonomous Driving via Pruning and Knowledge Distillation
- 基于梯度冲突惩罚的通道剪枝，保留关键特征。
- 蒸馏骨干与编码器中间特征，减少剪枝后性能损失。
- 在BDD100K上压缩32.7%参数，仍保持32.7 FPS实时推理。
自动驾驶系统依赖全景感知联合完成目标检测、可行驶区域分割和车道线分割。尽管多任务学习能有效整合这些任务，但模型参数与复杂度的增加使其难以部署于车载设备。为此，我们提出一种结合任务感知安全剪枝与特征级知识蒸馏的多任务模型压缩框架。安全剪枝策略融合基于泰勒展开的通道重要性与梯度冲突惩罚，保留关键通道并移除冗余及冲突通道。为缓解剪枝后的性能下降，进一步设计了不依赖任务头的知识蒸馏方法，将教师模型的中间骨干与编码器特征作为学生模型的指导。在BDD100K数据集上的实验表明，压缩后的模型参数减少32.7%，分割性能几乎无损，检测性能仅轻微下降（召回率-1.2%，mAP50 -1.8%），且仍可实现32.7 FPS的实时运行。结果表明，剪枝与知识蒸馏结合为多任务全景感知提供了有效的压缩方案。
原文摘要 · Abstract (English)
Autonomous driving systems rely on panoptic perception to jointly handle object detection, drivable area segmentation, and lane line segmentation. Although multi-task learning is an effective way to integrate these tasks, its increasing model parameters and complexity make deployment on on-board devices difficult. To address this challenge, we propose a multi-task model compression framework that combines task-aware safe pruning with feature-level knowledge distillation. Our safe pruning strategy integrates Taylor-based channel importance with gradient conflict penalty to keep important channels while removing redundant and conflicting channels. To mitigate performance degradation after pruning, we further design a task head-agnostic distillation method that transfers intermediate backbone and encoder features from a teacher to a student model as guidance. Experiments on the BDD100K dataset demonstrate that our compressed model achieves a 32.7% reduction in parameters while segmentation performance shows negligible accuracy loss and only a minor decrease in detection (-1.2% for Recall and -1.8% for mAP50) compared to the teacher. The compressed model still runs at 32.7 FPS in real-time. These results show that combining pruning and knowledge distillation provides an effective compression solution for multi-task panoptic perception.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。