Skip to main content
:fellow professors
:fellow professors
:all professors

Michele Magno

Fellow Professor of Embedded Systems & Edge AI 

title
PD Dr. 
email
location
ETH Zürich 
:back to overview

Michele Magno

Role at IT:U  

As a Fellow Professor of Embedded Systems & Edge AI at IT:U, Michele Magno contributes to the development of new academic initiatives within the IT:U Smart Space Sensing and Systems Lab (S³ Lab) and the IT:U Satellite Lab. His involvement focuses on strategic guidance in teaching and research areas such as Project-based Learning, EdgeAI, Intelligent Sensors and Systems, Wireless Sensors Networks, Energy-efficient  IoT and embedded systems, Battery-operated devices, Energy Harvesting for Smart Sensors, Autonomous Robots.

Research group:

The Smart Sensing and Systems Laboratory (S³ Lab) is a research group dedicated to advancing knowledge and technological innovation on sensing and physical-AI, with a strong commitment to teaching and cutting-edge research. Our research activities focus on the intersection of efficient intelligent systems and sustainable sensing technologies, exploring innovative solutions for future challenges and real-world applications. Our teaching and research efforts focus on areas such as Project-based Learning, Edge AI, Intelligent Sensors and Systems, Wireless Sensors Networks, Energy-efficient IoT and embedded systems, Battery-operated devices, Energy Harvesting for Smart Sensors, CubeSats and Sensing for earth observation, and Autonomous Robots. 

Smart Sensing and Systems Laboratory (S³ Lab)

:accepting applications
:read more

Short bio:

Michele Magno is a Privatdozent at ETH Zurich, Switzerland, where he is the Head of the Project-Based Learning Centre and lead a research group on Intelligence Sensors and Systems at ETH Zurich. He is a Fellow Professor at the Interdisciplinary Transformation University of Austria (IT:U) leading Smart Sensing and Systems Laboratory (S³ Lab). He has collaborated with several universities and research centers both academic and industrial, such as Tyndall Institute Ireland, Imperial College London, IBM Resarch, STMicroelectronics among others.   

He has published over 350 articles in international journals and conferences for which he received multiple best papers and poster awards, and he is IEEE Senior members since 2013.  

“Technology education must evolve with technology itself. Through Physical AI and project-based learning, we merge research and teaching, enabling students to design, build, and understand intelligent systems that interact with the real world. This fusion of edge intelligence and experiential learning prepares a new generation of innovators ready to tackle societal challenges with creativity, responsibility, and purpose”. 

Michele Magno

Publications

TitleYear
End-to-end multicamera event-image stitching and object detection on the edge
2025
Julian Moosmann, Jakub Mandula, Jiayong Li, Philipp Mayer, Luca Benini, Michele Magno
Target and Background Signatures XI: Traditional Methods and Artificial Intelligence 13…
:read more
Authors:Julian Moosmann, Jakub Mandula, Jiayong Li, Philipp Mayer, Luca Benini, Michele Magno
Publication date:10/28/2025
Journal:Target and Background Signatures XI: Traditional Methods and Artificial Intelligence
Volume:13673
Pages:225-235
Publisher:SPIE
Description:Wide Field-of-View (FoV) and high-resolution imaging are crucial capabilities for visual surveillance. However, they impose significant computational demands, posing significant challenges for embedded platforms targeting real-time performance. Event-based cameras help in curtailing computational effort by capturing only brightness changes at the pixel level, thereby reducing data volume while enhancing temporal resolution and dynamic range. Moreover, their inherent sensitivity to object edges improves the visibility of camouflage patterns. This work presents a real-time, wide-FoV event-based vision system based on a dual-camera setup with onboard object detection and tracking. The proposed system implements a low-latency end-to-end pipeline encompassing data capture, event-stream processing, image stitching, object detection, and tracking. Custom CUDA kernels are developed for efficient event …
:view less
QuantSparse: Comprehensively Compressing Video Diffusion Transformer with Model Quantization and Attention Sparsification
2025
Weilun Feng, Chuanguang Yang, Haotong Qin, Mingqiang Wu, Yuqi Li, Xiangqi Li, Zhulin An…
arXiv preprint arXiv:2509.23681
:read more
Authors:Weilun Feng, Chuanguang Yang, Haotong Qin, Mingqiang Wu, Yuqi Li, Xiangqi Li, Zhulin An, Libo Huang, Yulun Zhang, Michele Magno, Yongjun Xu
Publication date:09/28/2025
Journal:arXiv preprint arXiv:2509.23681
Description:Diffusion transformers exhibit remarkable video generation capability, yet their prohibitive computational and memory costs hinder practical deployment. Model quantization and attention sparsification are two promising directions for compression, but each alone suffers severe performance degradation under aggressive compression. Combining them promises compounded efficiency gains, but naive integration is ineffective. The sparsity-induced information loss exacerbates quantization noise, leading to amplified attention shifts. To address this, we propose textbf{QuantSparse}, a unified framework that integrates model quantization with attention sparsification. Specifically, we introduce textit{Multi-Scale Salient Attention Distillation}, which leverages both global structural guidance and local salient supervision to mitigate quantization-induced bias. In addition, we develop textit{Second-Order Sparse Attention Reparameterization}, which exploits the temporal stability of second-order residuals to efficiently recover information lost under sparsity. Experiments on HunyuanVideo-13B demonstrate that QuantSparse achieves 20.88 PSNR, substantially outperforming the state-of-the-art quantization baseline Q-VDiT (16.85 PSNR), while simultaneously delivering a textbf{3.68} reduction in storage and textbf{1.88} acceleration in end-to-end inference. Our code will be released in https://github.com/wlfeng0509/QuantSparse.
Resources:[PDF] from arxiv.org
:view less
:all publications

Positions offered

Communications & Efficiency: “Energy-Efficient Communication Systems for Next-Generation CubeSats”

Positions
1 PhD position

Application deadline:
January 31, 2026

:read more
:apply now

CubeSat Design & Sensing: “Embedded Sensors and CubeSat Architectures for Earth Observation Missions”

Positions
1 PhD position

Application deadline:
January 31, 2026

:read more
:apply now
:back to overview