DSORT-MCU: Detecting Small Objects in Real-Time on Microcontroller Units
Liam Boyle, Julian Moosmann, Nicolas Baumann, Seonyeong Heo, Michele, Magno

TL;DR
This paper introduces an adaptive tiling method for lightweight neural networks that enables accurate small object detection on low-power MCUs, achieving high F1-scores and low latency without sacrificing energy efficiency.
Contribution
It proposes a novel tiling approach for lightweight detection networks, improving small object detection accuracy on resource-constrained microcontrollers.
Findings
Boosts F1-score by up to 225% for FOMO and TinyissimoYOLO.
Reduces object count error by up to 89%.
Demonstrates real-time detection with low latency and energy use on RISC-V MCU.
Abstract
Advances in lightweight neural networks have revolutionized computer vision in a broad range of IoT applications, encompassing remote monitoring and process automation. However, the detection of small objects, which is crucial for many of these applications, remains an underexplored area in current computer vision research, particularly for low-power embedded devices that host resource-constrained processors. To address said gap, this paper proposes an adaptive tiling method for lightweight and energy-efficient object detection networks, including YOLO-based models and the popular FOMO network. The proposed tiling enables object detection on low-power MCUs with no compromise on accuracy compared to large-scale detection models. The benefit of the proposed method is demonstrated by applying it to FOMO and TinyissimoYOLO networks on a novel RISC-V-based MCU with built-in ML accelerators.…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
