Mohammad Loni is a researcher (Licentiate Degree) at the School of Innovation, Design, and Engineering at Mälardalen University since October 2017. He received his B.Sc. degree in computer hardware engineering and the M.Sc. degree in computer science from Shiraz University in 2017. He had collaboration in domestic industrial projects focusing on FPGA based radar emulators and embedded signal processing.

Currently, he is a member of the Dependable Platforms for Autonomous Systems and Control (DPAC), Deep Learning Accelerator on Commercial Programmable Devices (DeepMaker), and Fast and Sustainable Analysis Techniques for Advanced Real-Time Systems (FAST-ARTS) projects at Mälardalen University. He is mainly working on energy efficient designing neural network architecture for embedded systems.

My research interests are including deep learning, automated machine learning (AutoML), and heterogeneous embedded systems.

[Show all publications]

[Google Scholar author page]

Latest publications:

TAS:Ternarized Neural Architecture Search for Resource-Constrained Edge Devices (Mar 2022)
Mohammad Loni, Hamid Mousavi , Mohammad Riazati, Masoud Daneshtalab, Mikael Sjödin
Design, Automation and Test in Europe Conference (DATE'22)

FastStereoNet: A Fast Neural Architecture Search for Improving the Inference of Disparity Estimation on Resource-Limited Platforms (Oct 2021)
Mohammad Loni, Ali Zoljodi, Amin Majd , Byung Hoon Ahn , Masoud Daneshtalab, Mikael Sjödin, Hadi Esmaeilzadeh
IEEE Transactions on Systems, Man, and Cybernetics: Systems (SMCS)

Image Synthesisation and Data Augmentation for Safe Object Detection in Aircraft Auto-Landing System (Feb 2021)
Najda Vidimlic , Alexandra Levin , Mohammad Loni, Masoud Daneshtalab
16th International Conference on Computer Vision Theory and Applications (VISAPP 2021)

DeepMaker: Customizing the Architecture of Convolutional Neural Networks for Resource-Constrained Platforms (Dec 2020)
Mohammad Loni

Improving Motion Safety and Efficiency of Intelligent Autonomous Swarm of Drones (Aug 2020)
Amin Majd , Mohammad Loni, Golnaz Sahebi , Masoud Daneshtalab
Drones (Drones)

DenseDisp: Resource-Aware Disparity Map Estimation by Compressing Siamese Neural Architecture (Jul 2020)
Mohammad Loni, Ali Zoljodi, Amin Majd , Masoud Daneshtalab, Mikael Sjödin, Ben Juurlink , Reza Akbari
IEEE WORLD CONGRESS ON COMPUTATIONAL INTELLIGENCE (WCCI) 2020 (IEEE WCCI)

Project Title	Status
AutoDeep: Automatic Design of Safe, High-Performance and Compact Deep Learning Models for Autonomous Vehicles	active
DeepMaker: Deep Learning Accelerator on Commercial Programmable Devices	finished
DPAC - Dependable Platforms for Autonomous systems and Control	active
FAST-ARTS: Fast and Sustainable Analysis Techniques for Advanced Real-Time Systems	finished
GreenDL: Green Deep Learning for Edge Devices	active
HERO: Heterogeneous systems - software-hardware integration	active

MSc theses supervised (or examined):

Thesis Title	Status
Developing the First Carbon Footprint Indicator for Multi-modal Transport	available
Development of the First of its Kind Public Transport Crowdedness Indicator	available
Efficient Implementation of Ternary Neural Networks on FPGA	available
Energy Efficient Designing a Multi-task Deep Neural Network for Embedded Syetems	available
A Framework for Testing Redundant Components In Software and Hardware	in progress

We are using cookies to give you the best experience on our site. Cookies are files stored in your browser and are used by most websites to help personalise your web experience.

By continuing to use our website without changing the settings, you are agreeing to our use of cookies.

close

Mälardalen University

|

Box 325, 631 05 Eskilstuna

|

Box 883, 721 23 Västerås

|

+46-21-10 13 00, +46-16-15 36 00