Soroush Saryazdi

I am a computer science master's student at Concordia University, advised by Sudhir P. Mudur. Since January 2020, I have been collaborating with Robotics and Embodied AI Lab (REAL) at Mila, led by Liam Paull.

I'm broadly interested in computer vision and machine learning. My master's thesis is at the intersection of computer vision, deep learning, graphics, and robotics, where I focus on using end-to-end trainable models for recovering the 3D shape and/or materials of a scene. I am also interested in analyzing deep neural network components and architectures to get a better understanding of the learning process.

My master's research is partly funded by the Concordia Merit Scholarship.

Email  /  Twitter  /  Google Scholar  /  Github  /  LinkedIn  /  CV

profile photo
News
April 2021 Our paper titled "Disentangled Rendering Loss for Supervised Material Property Recovery" won the Best Student Paper Award at GRAPP 2021.
Nov.
2020
We released gradslam: an open-source library for differentiable dense SLAM.
Featured Papers

Please see my Google Scholar for a complete list of papers.

Disentangled Rendering Loss for Supervised Material Property Recovery
Soroush Saryazdi, Christian Murphy, Sudhir P. Mudur
GRAPP 2021 (Best Student Paper Award)
paper / cite

The disentangled rendering loss can be used in supervised material appearance estimation tasks for recovering more accurate individual BRDF parameters.

SLAM: Automagically differentiable SLAM
Krishna Murthy Jatavallabhula*, Soroush Saryazdi*, Ganesh Iyer, Liam Paull
CVPR workshop 2020
project page / code / paper / cite

SLAM (gradSLAM) is a differentiable computational graph view of dense SLAM which allows gradients from SLAM outputs to be backpropagated to the input.

The Problem of Entangled Material Properties in SVBRDF Recovery
Soroush Saryazdi, Christian Murphy, Sudhir P. Mudur
Eurographics workshop 2020
paper / cite

We analyze the landscape of the rendering loss and how it can negatively impact the network into estimating less accurate individual BRDF parameters.


Template source code from Jon Barron's website