在100KB内存下实现持续目标检测，自适应压缩提升性能不遗忘。
AHC: Meta-Learned Adaptive Compression for Continual Object Detection on Memory-Constrained Microcontrollers
- 用元学习快速适配新任务，5步内完成压缩参数更新。
- 分层多尺度压缩匹配特征图冗余，比率分别设为8:1、6.4:1、4:1。
- 双存储架构+重要性整合，100KB内有效防止灾难性遗忘。
在内存不足100KB的微控制器上部署持续目标检测，需高效自适应特征压缩。现有方法依赖固定压缩策略（如FiLM调节），无法适配不同任务特性，导致内存利用不足与灾难性遗忘。本文提出自适应分层压缩（AHC），包含三项创新：(1) 基于MAML的真实梯度压缩，可在5次内循环步骤内快速适配新任务；(2) 分层多尺度压缩，按比例8:1（P3）、6.4:1（P4）、4:1（P5）匹配FPN冗余模式；(3) 双记忆架构，结合短期与长期特征库，基于重要性进行固化，在硬性100KB预算下运行。理论证明灾难性遗忘上界为O(ε√T + 1/√M)，其中ε为压缩误差，T为任务数，M为内存大小。在CORe50、TiROD和PASCAL VOC上，对比微调、EWC、iCaRL三个基线，AHC在100KB回放预算下实现可实用的持续检测，通过均值池化压缩特征重放结合EWC正则与特征蒸馏，达到有竞争力的准确率。
原文摘要 · Abstract (English)
Deploying continual object detection on microcontrollers (MCUs) with under 100KB memory requires efficient feature compression that can adapt to evolving task distributions. Existing approaches rely on fixed compression strategies (e.g., FiLM conditioning) that cannot adapt to heterogeneous task characteristics, leading to suboptimal memory utilization and catastrophic forgetting. We introduce Adaptive Hierarchical Compression (AHC), a meta-learning framework featuring three key innovations: (1) true MAML-based compression that adapts via gradient descent to each new task in just 5 inner-loop steps, (2) hierarchical multi-scale compression with scale-aware ratios (8:1 for P3, 6.4:1 for P4, 4:1 for P5) matching FPN redundancy patterns, and (3) a dual-memory architecture combining short-term and long-term banks with importance-based consolidation under a hard 100KB budget. We provide formal theoretical guarantees bounding catastrophic forgetting as O(ε{sq.root(T)} + 1/{sq.root(M)}) where ε is compression error, T is task count, and M is memory size. Experiments on CORe50, TiROD, and PASCAL VOC benchmarks with three standard baselines (Fine-tuning,EWC, iCaRL) demonstrate that AHC enables practical continual detection within a 100KB replay budget, achieving competitive accuracy through mean-pooled compressed feature replay combined with EWC regularization and feature distillation.
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。