Suryansh
 
3D Computer Vision
ETH Zurich
Computer Vision Lab
Dept. of Information Technology and Electrical Engineering
Résumé (updated February 2020)
sukumar at vision dot ee.ethz.ch
     

I am a computer science researcher primarily focused in the areas of 3D computer vision. Presently, I hold the position of Professorship for Computer Vision in the Dept. of Information Technology and Electrical Engineering at ETH Zürich, where I am advised by Prof. Dr. Luc Van Gool. I received Ph.D. in Engineering and Computer Science from Australian National University on the topic Non-Rigid Structure from Motion. As a researcher, I am fascinated by how mathematical construction can precisely represent the perceptual concepts of images such as 3D scene geometry, motions, lights and color. Therefore, I aim to use these mathematical concepts to enable machines for a broader adoption using visual data. Presently, I work on the projects that are supported by ETH Zürich Foundation and Google Zurich.

Besides my work life, I like to watch movies, musicals, learn to cook food, photograhy and, love to travel new places.

Research Interests

Geometric Computer Vision, Mathematical Optimization, Abstract Algebra, Geometric AI and Robotics. From time to time, I also investigate topics such as graph theory, topological manifolds and compressed sensing.

Recent News

Jun. 2020
Project Page for Non-Rigid Structure from Motion is now online.
Mar. 2020
One of our research work got accepted for publication in WACV 2020.
Feb. 2020
Nominated for J. G. Crawford Prize at ANU for Best Ph.D. Thesis 2019.
Dec. 2019
One of our research work got accepted for publication in T-PAMI 2019.
Jun. 2019
Invited talk at Dynavis CVPR 2019 to present my work MoJu.
May 2019
Moved to Google Research, New York USA to work on Geometric Learning
Feb. 2019
One of our research work got accepted for publication in CVPR 2019.
Jan. 2019
Invited talk at Samsung Research America, California.
Feb. 2018
One of our research work got accepted for publication in CVPR 2018.
Oct. 2017
One of our research work got accepted for publication in ICCV 2017.
July 2017
Best Algorithm Award from Disney Research for our work "Multi-body NRSfM" at CVPR 2017.

Recent Awards and Achievements

  • Nominated for J. G. Crawford Prize at ANU for Best Ph.D. Thesis 2019.
  • Australian National University Vice-Chancellor Grant Award.
  • Best Algorithm Award in CVPR NRSFM Challenge 2017 by Disney Research.
  • ANU-HDR Merit Scholarship Student Award, funded in part by Australian Research Council.

Students

ETHZ
Erik Sandström
  Ph.D. Student (CVL)
  ETH Zurich (Nov' 19-)
  Area: 3D Vision.
ETHZ
Berk Kaya
  Ph.D. Student (CVL)
  ETH Zurich (Nov' 19-)
  Area: 3D vision.
ETHZ
Erik Endsjo Goron
  MS Student (CVL)
  ETH Zurich (Mar. 20-)
  Topic: Video Clustering.
ETHZ
Rhea Sukthanker
  MS Student (CVL)
  ETH Zurich (Mar. 20-)
  Topic: Video Clustering.

Advisor and Co-Advisors

ETHZ
Luc Van Gool
  Advisor, CVL Lab Head
  ETH Zurich, KU Leuven
  Professor
ETHZ
Vittorio Ferrari
  Co-Advisor
  Google Research Zurich
  Research Scientist
ETHZ
Cristian Sminchisescu
  Co-Advisor
  Google Research Zurich
  Research Scientist
ETHZ
Zhiwu Huang
  Co-Advisor
  ETH Zurich
  Researcher

Recent Research Projects and Highlights

Complex Surface 3D Reconstruction
NRSFM
Dense Non-Rigid Structure from Motion
DNRSFM
Non-Euclidean Geometry and Optimization
MBNRSFM
RGB-D Reconstruction of Dynamic Scene
Complex

Classical Non-Rigid Structure from Motion
NRSFM
Inverse Problems*
Inverse
Dense 3D reconstruction of a Dynamic Scene
Complex
Multi-body Non-Rigid Structure from Motion
MBNRSFM

Recent Readings and Courses

Courses:

  • Unsupervised Learning (Prof. Rene Vidal)
  • Convex Optimization (Prof. Stephen P. Boyd)
  • Linear Algebra (Prof. Gilbert Strang)
  • Differential Geometry

Books and Good Papers (to me) by different authors:

    Books
  • Visual Reconstruction (Andrew Blake, Andrew Zissermann)
  • Multiple View Geometry in Computer Vision (Richard Hartley, Andrew Zissermann)
  • Numerical Geometry of Non-Rigid Shapes (A. Bronstein, M. Bronstein)
  • Optimization Algorithm on Matrix Manifolds (P.A. Absil, R. Mahony)
  • Generalized Principal Component Analysis (R. Vidal, Y. Ma, and S. Sastry)
    • Papers
  • Long short-term memory, Neural computation 1997.
  • A Theory of Shape by Space Carving, IJCV 2000.
  • Intrinsic Scene Properties from a Single RGB-D Image, CVPR 2013.
  • Rotation Averaging, IJCV 2013.
  • Convergent Tree-reweighted Message Passing for Energy Minimization, T-PAMI, 2006.
  • Multiple-View Geometry Under the l-infinity norm, T-PAMI (2008).
  • A Simple Prior Free Factorization for Non-Rigid Structure from Motion, IJCV (2014).
  • 3D Scene Flow Estimation with Piecewise Rigid Scene Model, IJCV (2015).
  • Robust relative rotation averaging, T-PAMI (2018).
  • An Interior-Point Method for Large-Scale l1-Regularized Least Squares, IEEE Journal of Selected Topics in Signal Processing, 2007
Some of the best results of all time (From the artists of Mathematics, Ramanujan, Euler, C. F. Gauss and H. Grassmann. Ramanujan's Pi Formula, Euler's Identity, DFT, Grassmann Algebra).
  • Ramanujan's Pi Formula: PI
  • Euler's Identity: Euler
  • Discrete Fourier Transform: DFT
  • Hermann Grassmann: The notion of subspace, independence, linear combination, span, dimension, projection of elements onto subspaces.


*The image shown for Inverse Problems is taken from Maks Ovsjanikov presentation slides.

Note: Please be advised that the simulation results, research publications and images present on this webpage are subjected to copyright. The copyright of the contents are retained by the respective author/author's.