MCUBench: A Benchmark of Tiny Object Detectors on MCUs

TL;DR

MCUBench is a comprehensive benchmark evaluating over 100 YOLO-based object detection models on MCUs, providing detailed performance metrics to aid in model selection under resource constraints.

Contribution

Introduces MCUBench, a detailed benchmark for tiny object detectors on MCUs, with a controlled comparison and Pareto analysis of model efficiency.

Findings

Modern detection heads improve efficiency.

Legacy models like YOLOv3 can achieve competitive tradeoffs.

Benchmark aids in selecting models based on resource constraints.

Abstract

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.