DeepMaker: Deep Learning Accelerator on Commercial Programmable Devices



Start date:


End date:


DeepMaker aims to provide a framework to generate synthesizable accelerators of Deep Neural Networks (DNNs) that can be used for different FPGA fabrics. DeepMaker enables effective use of DNN acceleration in commercially available devices that can accelerate a wide range of applications without a need of costly FPGA reconfigurations. 

[Show all publications]

An energy-efficient partition-based XYZ-planar routing algorithm for a wireless network-on-chip (Dec 2018)
Fahimeh Yazdanpanah , Raheel Afsharmazayejani , Amin Rezaei , Masoud Daneshtalab
The Journal of Supercomputing (Supercomputing)

Designing Compact Convolutional Neural Network for Embedded Stereo Vision Systems (Sep 2018)
Mohammad Loni, Amin Majd , Abdolah Loni , Masoud Daneshtalab, Mikael Sjödin, Elena Troubitsyna
IEEE 12th International Symposium on Embedded Multicore/Many-core Systems-on-Chip (MCSoC-2018)

ADONN: Adaptive Design of Optimized Deep Neural Networks for Embedded Systems (Aug 2018)
Mohammad Loni, Masoud Daneshtalab, Mikael Sjödin
21st Euromicro Conference on Digital System Design (DSD'18)

Using Optimization, Learning, and Drone Reflexes to Maximize Safety of Swarms of Drones (Jul 2018)
Amin Majd , Adnan Ashraf , Elena Troubitsyna , Masoud Daneshtalab
IEEE Congress on Evolutionary Computation (IEEE CEC'18)

A Novel Two-Step Method for Stereo Vision Algorithm to Reduce Search Space (Jun 2018)
Mehdi Kokhazadeh , Zahra Kokhazadeh , Masoud Daneshtalab
International Conference on electrical Engineering (ICEE)

HoneyWiN: Novel Honeycomb-Based Wireless NoC Architecture in Many-Core Era (May 2018)
Raheel Afsharmazayejani , Fahimeh Yazdanpanah , Masoud Daneshtalab
International Symposium on Applied Reconfigurable Computing (ARC)

Saab AB, Avionics Systems Industrial
Unibap AB Industrial

Masoud Daneshtalab, Associate Professor,Docent

Phone: +46-(0)736620918