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.
@article{arxiv.2409.18866,
title = {MCUBench: A Benchmark of Tiny Object Detectors on MCUs},
author = {Sudhakar Sah and Darshan C. Ganji and Matteo Grimaldi and Ravish Kumar and Alexander Hoffman and Honnesh Rohmetra and Ehsan Saboori},
journal= {arXiv preprint arXiv:2409.18866},
year = {2024}
}
Comments
Code and data are available at https://github.com/Deeplite/deeplite-torch-zoo