MCUBench: A Benchmark of Tiny Object Detectors on MCUs
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910623672041472 |
|---|---|
| author | Sah, Sudhakar Ganji, Darshan C. Grimaldi, Matteo Kumar, Ravish Hoffman, Alexander Rohmetra, Honnesh Saboori, Ehsan |
| author_facet | Sah, Sudhakar Ganji, Darshan C. Grimaldi, Matteo Kumar, Ravish Hoffman, Alexander Rohmetra, Honnesh Saboori, Ehsan |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_18866 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MCUBench: A Benchmark of Tiny Object Detectors on MCUs Sah, Sudhakar Ganji, Darshan C. Grimaldi, Matteo Kumar, Ravish Hoffman, Alexander Rohmetra, Honnesh Saboori, Ehsan Computer Vision and Pattern Recognition 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. |
| title | MCUBench: A Benchmark of Tiny Object Detectors on MCUs |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2409.18866 |