在微控制器上高效搜索目标检测模型，兼顾精度与速度。
ELASTIC: Efficient Once For All Iterative Search for Object Detection on Microcontrollers
- 分模块迭代优化，通过种群传递机制加速搜索
- 在PascalVOC上达72.3% mAP，比MCUNET高20.9%
- 适合资源受限设备部署，能显著降低能耗和延迟
在微型机器学习平台部署高性能目标检测器面临硬件资源紧张与检测流水线模块复杂性的双重挑战。现有神经架构搜索方法或仅优化单一模块，牺牲跨模块协同；或需全局搜索，计算成本过高。本文提出ELASTIC（高效一次为所有微控制器上的目标检测迭代搜索），一种统一且硬件感知的NAS框架，通过循环交替优化主干、颈部和头部等模块。ELASTIC引入新型种群传递机制，在进化搜索中保留高质量候选，实现更快收敛，最多提升8%最终mAP，消除无种群传递时的搜索不稳定性。实测表明，在相同搜索预算下，相比渐进式NAS策略，ELASTIC在SVHN上实现+4.75% mAP与2倍加速；在PascalVOC上提升9.09% mAP。ELASTIC在PascalVOC上达到72.3% mAP，较MCUNET高出20.9%，较TinyissimoYOLO高16.3%。在MAX78000/MAX78002微控制器上部署时，其模型优于Analog Devices的TinySSD基线，最高节能71.6%，延迟降低2.4倍，且在多个数据集上mAP提升最高达6.99个百分点。实验视频与代码已开源。
原文摘要 · Abstract (English)
Deploying high-performance object detectors on TinyML platforms poses significant challenges due to tight hardware constraints and the modular complexity of modern detection pipelines. Neural Architecture Search (NAS) offers a path toward automation, but existing methods either restrict optimization to individual modules, sacrificing cross-module synergy, or require global searches that are computationally intractable. We propose ELASTIC (Efficient Once for AlL IterAtive Search for ObjecT DetectIon on MiCrocontrollers), a unified, hardware-aware NAS framework that alternates optimization across modules (e.g., backbone, neck, and head) in a cyclic fashion. ELASTIC introduces a novel Population Passthrough mechanism in evolutionary search that retains high-quality candidates between search stages, yielding faster convergence, up to an 8% final mAP gain, and eliminates search instability observed without population passthrough. In a controlled comparison, empirical results show ELASTIC achieves +4.75% higher mAP and 2x faster convergence than progressive NAS strategies on SVHN, and delivers a +9.09% mAP improvement on PascalVOC given the same search budget. ELASTIC achieves 72.3% mAP on PascalVOC, outperforming MCUNET by 20.9% and TinyissimoYOLO by 16.3%. When deployed on MAX78000/MAX78002 microcontrollers, ELASTICderived models outperform Analog Devices' TinySSD baselines, reducing energy by up to 71.6%, lowering latency by up to 2.4x, and improving mAP by up to 6.99 percentage points across multiple datasets. The experimental videos and codes are available on the project website (https://nail-uh.github.io/elastic.github.io/).
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。