Amr A. Suleiman
I am a PhD candidate in the Department of Electrical Engineering and Computer Science at Massachusetts Institute of Technology. I received the B.E. degree in Electronics and Electrical Communications Engineering and the M.S degree in Electronics from Faculty of Engineering Cairo University in Egypt in 2008 and 2011 respectively. I also received the M.S. degree in Electrical Engineering and Computer Science from Massachusetts Institute of Technology in 2013. Currently I'm working towards my Ph.D. degree.
Research
My research supervisor is Prof. Vivienne Sze at the Research Lab of Electronics (RLE). I am part of the Energy Efficient Multimedia Systems group.
I'm interested in low power and high performance digital VLSI for signal processing in general. My work focuses on developing new energy-efficient hardware implementations for machine vision (e.g. detection, recognition, tracking). In addition to optimizations at the architecture and circuit level, system optimization is done at the algorithmic level to efficiently trade-off accuracy and hardware complexity. My recent chip was an object detector processing HD images at 30fps in real time and low power using the Deformable Parts Models (DPM).
Journals & Magazines Publications
-
A 58.6mW 30fps Real-Time Programmable Multi-Object Detection Accelerator with Deformable Parts Models on Full HD 1920×1080 Videos,
Amr Suleiman, Zhengdong Zhang, and Vivienne Sze, IEEE Journal of Solid State Circuits (JSSC), vol., no. 99, pp.1–12, 2017. [
PAPER]
- An Energy-Efficient Hardware Implementation of HOG-based Object Detection at 1080HD 60 fps with Multi-Scale Support, Amr Suleiman, and Vivienne Sze, Journal of Signal Processing Systems (SiPS Special Issue), 2015, [PAPER]
- Model Predictive Control Equalization for High-Speed I/O Links, Amr Suleiman, Ranko Sredojevic, and Vladimir Stojanovic, IEEE Transactions of Circuits and Systems I: Regular Papers, 2014. [PAPER]
Conferences Publications (* indicates authors with equal contributions)
- Visual-Inertial Odometry on Chip: An Algorithm-and-Hardware Co-design Approach, Zhengdong Zhang*, Amr Suleiman*, Luca Carlone, Vivienne Sze, and Sertac Karaman, To appear in Robotics: Science and Systems Conference (RSS), (accepted), July 2017.
- Closing the Energy Gap Between HOG and CNN Features for Embedded Vision, Amr Suleiman*, Yu-Hsin Chen*, Joel Emer, and Vivienne Sze, To appear in IEEE International symposium on Circuits and Systems (ISCAS), (accepted), May 2017. [PAPER]
-
Hardware for Machine Learning: Challenges and Opportunities, Vivienne Sze, Joel Emer, Yu-Hsin Chen,
Amr Suleiman and Zhengdong Zhang, To appear in IEEE Custom Integrated Circuits Conference (CICC), (accepted), April 2017. [
PAPER]
- A 58,6mW Real-Time Programmable Object Detector with Multi-Scale Multi-Object Support Using Deformable Parts Model on 1920x1080 Video at 30fps, Amr Suleiman, Zhengdong Zhang, and Vivienne Sze, IEEE Symposium on VLSI Circuits 2016, Honolulu, USA. [PAPER]
- Energy-Efficient HOG-based Object Detection at 1080HD 60 fps with Multi-Scale Support, Amr Suleiman, and Vivienne Sze, IEEE Workshop on Signal Processing Systems (SiPS) 2014, Belfast, UK. [PAPER]
- ASIC Implementation of Cairo University SPARC (CUSPARC) Embedded Processor, Amr Suleiman, Alhassan Khedr, and Serag Habib, IEEE International Conference on Microelectronics (ICM) 2010, Cairo, Egypt. [PAPER]
- CUSPARC IP Processor: Design, Characterization and Applications, Ezz Hussein, Shoukry Shams, Mohamed Ali, Amr Suleiman et al., IEEE International Conference on Microelectronics (ICM) 2010, Cairo, Egypt. [PAPER]
Awards and Honors
- Winner of the 2015 Broadcom Foundation University Research Competition - Third Place
- Full-tuition Research fellowship from the Egyptin Ministry of Finance, 2011 - present
- The best graduation project in electronics and communication in the IEEE Egyptian Engineering Day 2008. (Project: ASIC implementation of CUSPARC processor)
Contact
Energy Efficient Multimedia Systems Group
77 Massachusetts Avenuw, 38-266
Cambridge, MA 02139
suleiman@mit.edu