Note: I have moved to ASU. Please navigate to my new group website here

Suren Jayasuriya

In January 2018, I will be an assistant professor at Arizona State University. Currently, I am a postdoctoral fellow at the Robotics Institute at Carnegie Mellon University. I received my Ph.D. in ECE at Cornell University in Jan 2017, under the direction of Dr. Alyosha Molnar. Before Cornell, I graduated from the University of Pittsburgh in 2012 with a B.S. in Mathematics (with departmental honors) and a B.A. in Philosophy.

Email  /  CV  /  PhD Thesis


In January 2018, I will be joining Arizona State University as an assistant professor in Arts, Media and Engineering (AME) and Electrical Engineering (ECEE). I am actively looking for PhD students to work with, feel free to send me an email along with your CV/resume! Here are copies of my research statement and teaching statement.

Research Interests

I am interested in building new types of computational cameras, systems, and visual computing algorithms that can extract and understand more information from the world around us. This research combines techniques in sensor hardware, computer vision and graphics, computer architecture and systems, machine learning, optics, and signal processing to create futuristic imaging systems and applications. Feel free to check out my publications below.


Reconfiguring the Imaging Pipeline for Computer Vision
Mark Buckler, Suren Jayasuriya, and Adrian Sampson
International Conference on Computer Vision (ICCV), Oct 2017
Paper Github

We reexamine the ISP pipeline to see which stages are necessary for computer vision and deep learning, and propose a new image sensor + ISP design to save energy for embedded computer vision applications.

Compressive Light Field Reconstructions using Deep Learning
Mayank Gupta*, Arjun Jauhari*, Kuldeep Kulkarni, Suren Jayasuriya, Alyosha Molnar, and Pavan Turaga
CVPR Workshop on Computational Cameras and Displays (CCD) , July 2017

We recover 4D light fields from compressively sensed 2D measurements using deep learning at faster times than previous methods. [* = equal authorship]

Reconstructing Intensity Images from Binary Spatial Gradient Cameras
Suren Jayasuriya, Orazio Gallo, Jinwei Gu, Timo Aila, and Jan Kautz
CVPR Workshop on Embedded Vision (EVW), July 2017

We use a deep learning autoencoder to recover intensity images from binary spatial gradient cameras.

Experiences using a Novel Python-Based Hardware Modeling Framework for Computer Architecture Test Chips
Christopher Torng, Moyang Wang, Bharath Sudheendra, Nagaraj Murali, Suren Jayasuriya, Shreesha Srinath, Taylor Pritchard, Robin Ying, and Christopher Batten
28th ACM/IEEE Symp. on High-Performance Chips (HOTCHIPS'16) Student Poster Session (Technical Abstract), August 2016

We describe a toolflow that uses PyMTL, a hardware modeling language, to facilitate vertically-oriented computer architecture research and agile hardware design. In particular, we taped out a RISC processor with embedded SRAM and HLS-generated accelerator as a proof of concept.

ASP Vision: Optically Computing the First Layer of CNNs using Angle Sensitive Pixels
Huaijin Chen*, Suren Jayasuriya*, Jiyue Yang, Judy Stephen, Sriram Sivaramakrishnan, Ashok Veeraraghavan, Alyosha Molnar
IEEE Computer Vision and Pattern Recognition (CVPR), 2016

We use ASPs to optically compute the first layer of a convolutional neural network, saving energy/bandwidth without sacrificing visual recognition performance. (oral presentation, less than 4% acceptance rate) . [* = equal authorship]

Depth Fields: Extending Light Field Techniques to Time-of-Flight Imaging
Suren Jayasuriya, Adithya Pediredla, Sriram Sivaramakrishnan, Alyosha Molnar, Ashok Veeraraghavan
International Conference on 3D Vision (3DV), 2015

We combine light field and time-of-flight imaging for robust depth mapping applications including imaging past partial and scattering occluders, phase unwrapping, and synthetic aperture refocusing. (oral presentation)

Dual Light Field and Polarization Imaging using Diffractive CMOS Image Sensors
Suren Jayasuriya, Sriram Sivaramakrishnan, Ellen Chuang, Debashree Guruaribam, Albert Wang, Alyosha Molnar
Optics Letters, 2015

We characterize the polarization response of ASPs, and show applications to imaging stress-induced birefringence and removing specular highlights from light field depth maps.

A Baseband Technique for Automated LO Suppression Achieving less than -80dBm in Wideband Passive Mixer-first Receivers
Suren Jayasuriya*, Dong Yang*, Alyosha Molnar
IEEE Custom Integrated Circuits Conference (CICC), 2014

We show a RF technique to suppress LO leakage on mixer-first receivers. [* = equal authorship]

A Switchable Light Field Camera Architecture using Angle Sensitive Pixels and Dictionary-based Sparse Coding
Matthew Hirsch*, Sriram Sivaramakrishnan*, Suren Jayasuriya*, Albert Wang, Alyosha Molnar, Ramesh Raskar, Gordon Wetzstein
IEEE International Conference on Computational Photography (ICCP), 2014

We show how dictionary-based sparse coding can be used to recover high resolution 4D light fields from ASPs. (Best paper award at ICCP 2014!) [* = equal authorship]

Changing Cycle Lengths in State-Transition Models: Challenges and Solutions
Jagpreet Chhatwal*, Suren Jayasuriya*, Elamin Elbasha
Medical Decision Making, July 2016

We explore how Markov transition models can be adapted to different cycle lengths (such as yearly to monthly), and outline under what conditions is it possible. [* = equal authorship]

Changing Cycle Lengths in State-Transition Models: Doing it the Right Way
Jagpreet Chhatwal, Suren Jayasuriya, Elamin Elbasha
International Society for Pharmacoeconomics and Outcomes Research, 2015

Invited paper for our Medical Decision Making journal paper on changing cycle lengths in Markov models.

On the Inverse Erdos-Heilbronn Problem for Restricted Set Addition in Finite Groups
Suren Jayasuriya, Steven Reich, Jeffrey Paul Wheeler
ArXiv, 2012

We prove some results on the nature of restricted sumsets which achieve the Erdos-Heilbronn lower bound.

Effects of Time-Dependent Stimuli in a Competitive Neural Network Model of Perceptual Rivalry
Suren Jayasuriya, Zachary Kilpatrick
Bulletin of Mathematical Biology, 2012

We show the effects of time-varying inputs to a coupled differential equations model of neural populations in the visual cortex for binocular rivalry.


2nd Place in Cornell's 3MT (Three Minute Thesis) Competition, Spring 2016

Qualcomm Innovation Fellowship, 2015-2016

Cornell ECE Outstanding PhD TA Award, 2015

Best Paper Award at ICCP 2014

NSF Graduate Research Fellowship, 2013-2016

Jacobs Fellowship - Cornell ECE, 2012-2013

Culver Award (awarded for achievement in mathematics), University of Pittsburgh, Spring 2011

University Scholar (top 2% of undergrads in Arts/Sciences), University of Pittsburgh, Spring 2011

Honors Tuition Scholarship, University of Pittsburgh, 2008-2012


ECE 4250 Digital Signal and Image Processing, Co-Instructor, Spring 2016

Lego Robotics Outreach at Cayuga Heights Elementary, Fall 2015

ECE 2100 Introduction to Circuits, TA, Spring 2014

Tutor at Math Assistance Center, University of Pittsburgh, Fall 2011-Spring 2012

Stat 1000 - Applied Statistical Methods (Honors), TA, Fall 2010

Students Advised: Judy Stephen (BS ’17), Grace Shih (BS ’16), Cyrus Moradi (BS ’16), Yang Li (MEng ’15), Ellen Chuang (MEng ’15), Debashree Guruaribam (MEng ’15), Einar Veizaga (MEng ’16), Jiyue Yang (BS ’16), Arjun Jauhari (MEng ’16)

This website template was graciously shared by Jonathan Barron