MCUBench: A Benchmark of Tiny Object Detectors on MCUs
This work provides a tool for benchmarking and selecting tiny object detectors on MCUs, addressing resource constraints in embedded systems, but it is incremental as it builds on existing YOLO models and datasets.
The authors introduced MCUBench, a benchmark evaluating over 100 YOLO-based object detection models on MCUs using the VOC dataset, providing metrics like mAP, latency, RAM, and Flash usage. Their Pareto-optimal analysis showed that integrating modern detection heads and training techniques enables efficient tradeoffs, with legacy models like YOLOv3 achieving competitive performance.
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.