评测100多个YOLO模型在MCU上的表现，帮开发者选高效检测模型。
MCUBench: A Benchmark of Tiny Object Detectors on MCUs
- 在7种MCU上测试上百个YOLO模型，统一训练流程确保公平比较。
- 发现现代检测头与训练技巧可让YOLOv3等旧模型实现高mAP与低延迟平衡。
- 提供精确的精度、延迟、内存占用数据，适合嵌入式部署场景参考。
我们提出MCUBench，一个包含超过100个基于YOLO的目标检测模型在七个不同MCU上对VOC数据集的评估基准。该基准提供了多种输入分辨率下，各类基于YOLO的一阶段检测器在平均精度（mAP）、延迟、RAM和Flash使用方面的详细数据。通过固定训练流程进行受控对比，我们收集了全面的性能指标。帕累托最优分析表明，整合现代检测头和训练技术，使包括YOLOv3等经典模型在内的多种YOLO架构，均能实现良好的mAP与延迟权衡。MCUBench为评估现代目标检测器在MCU上的表现提供了重要工具，并支持根据特定约束条件进行模型选择。
原文摘要 · Abstract (English)
We introduce MCUBench, a benchmark featuring over 100 YOLO-based object detection models evaluated on the VOC dataset across seven different MCUs. This benchmark provides detailed data on average precision, latency, RAM, and Flash usage for various input resolutions and YOLO-based one-stage detectors. By conducting a controlled comparison with a fixed training pipeline, we collect comprehensive performance metrics. Our Pareto-optimal analysis shows that integrating modern detection heads and training techniques allows various YOLO architectures, including legacy models like YOLOv3, to achieve a highly efficient tradeoff between mean Average Precision (mAP) and latency. MCUBench serves as a valuable tool for benchmarking the MCU performance of contemporary object detectors and aids in model selection based on specific constraints.
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。