STRESNET & STYOLO : A New Family of Compact Classification and Object Detection Models for MCUs

Sudhakar Sah, Ravish Kumar
Proceedings of the The 39th Canadian Conference on Artificial Intelligence, PMLR 318:38-51, 2026.

Abstract

Recent advancements in lightweight neural networks have significantly improved the efficiency of deploying deep learning models on edge hardware. However, most existing architectures still compromise accuracy for latency, which limits their applicability on MCU/NPU-based devices. In this work, we introduce two new model families — STResNet for image classification and STYOLO for object detection — jointly optimized for accuracy, efficiency, and memory footprint on resource-constrained platforms. The proposed STResNet series (ranging from Nano to Tiny variants) achieves competitive ImageNet-1K accuracy within a 4M parameter budget. Specifically, STResNetMilli attains 70.0% Top-1 accuracy with only 3.0M parameters, outperforming MobileNetV1 and ShuffleNetV2 at comparable computational complexity. For object detection, STYOLOMicro and STYOLOMilli achieve 30.5% and 33.6% mAP, respectively, on the MS-COCO dataset, surpassing YOLOv5n and YOLOX-Nano in both accuracy and efficiency. Furthermore, when STResNetMilli is used as a backbone with the Ultralytics detection head, it approaches the performance of the YOLOv11n model under the latest Ultralytics training environment.

Cite this Paper

BibTeX
@InProceedings{pmlr-v318-sah26a, title = {STRESNET & STYOLO : A New Family of Compact Classification and Object Detection Models for MCUs}, author = {Sah, Sudhakar and Kumar, Ravish}, booktitle = {Proceedings of the The 39th Canadian Conference on Artificial Intelligence}, pages = {38--51}, year = {2026}, editor = {Bouzar-Benlabiod, Lydia and Leung, Carson}, volume = {318}, series = {Proceedings of Machine Learning Research}, month = {25--29 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v318/main/assets/sah26a/sah26a.pdf}, url = {https://proceedings.mlr.press/v318/sah26a.html}, abstract = {Recent advancements in lightweight neural networks have significantly improved the efficiency of deploying deep learning models on edge hardware. However, most existing architectures still compromise accuracy for latency, which limits their applicability on MCU/NPU-based devices. In this work, we introduce two new model families — STResNet for image classification and STYOLO for object detection — jointly optimized for accuracy, efficiency, and memory footprint on resource-constrained platforms. The proposed STResNet series (ranging from Nano to Tiny variants) achieves competitive ImageNet-1K accuracy within a 4M parameter budget. Specifically, STResNetMilli attains 70.0% Top-1 accuracy with only 3.0M parameters, outperforming MobileNetV1 and ShuffleNetV2 at comparable computational complexity. For object detection, STYOLOMicro and STYOLOMilli achieve 30.5% and 33.6% mAP, respectively, on the MS-COCO dataset, surpassing YOLOv5n and YOLOX-Nano in both accuracy and efficiency. Furthermore, when STResNetMilli is used as a backbone with the Ultralytics detection head, it approaches the performance of the YOLOv11n model under the latest Ultralytics training environment.} }
Endnote
%0 Conference Paper %T STRESNET & STYOLO : A New Family of Compact Classification and Object Detection Models for MCUs %A Sudhakar Sah %A Ravish Kumar %B Proceedings of the The 39th Canadian Conference on Artificial Intelligence %C Proceedings of Machine Learning Research %D 2026 %E Lydia Bouzar-Benlabiod %E Carson Leung %F pmlr-v318-sah26a %I PMLR %P 38--51 %U https://proceedings.mlr.press/v318/sah26a.html %V 318 %X Recent advancements in lightweight neural networks have significantly improved the efficiency of deploying deep learning models on edge hardware. However, most existing architectures still compromise accuracy for latency, which limits their applicability on MCU/NPU-based devices. In this work, we introduce two new model families — STResNet for image classification and STYOLO for object detection — jointly optimized for accuracy, efficiency, and memory footprint on resource-constrained platforms. The proposed STResNet series (ranging from Nano to Tiny variants) achieves competitive ImageNet-1K accuracy within a 4M parameter budget. Specifically, STResNetMilli attains 70.0% Top-1 accuracy with only 3.0M parameters, outperforming MobileNetV1 and ShuffleNetV2 at comparable computational complexity. For object detection, STYOLOMicro and STYOLOMilli achieve 30.5% and 33.6% mAP, respectively, on the MS-COCO dataset, surpassing YOLOv5n and YOLOX-Nano in both accuracy and efficiency. Furthermore, when STResNetMilli is used as a backbone with the Ultralytics detection head, it approaches the performance of the YOLOv11n model under the latest Ultralytics training environment.
APA
Sah, S. & Kumar, R.. (2026). STRESNET & STYOLO : A New Family of Compact Classification and Object Detection Models for MCUs. Proceedings of the The 39th Canadian Conference on Artificial Intelligence, in Proceedings of Machine Learning Research 318:38-51 Available from https://proceedings.mlr.press/v318/sah26a.html.

Related Material