Rudra

Rudrasis Chakraborty
Postdoctoral Researcher
UC Berkeley/ ICSI

rudra@berkeley.edu
Google Scholar | CV | Github

About

I do research in the intersection of statistics, differential geometry and deep learning. I am a member of Dr. Stella Yu's group at ICSI. I am also a member of Berkeley Deep Drive, UC Berkeley. I was a member of the CVGMI group at University of Florida (UF).

I joined UF–Gainesville in 2013 and finished my PhD in April 2018, where I was advised by Baba Vemuri. After my PhD, I continued my research at UF as a postdoctoral associate for 6 months.

News

7/18 Submitted extended version of our paper on Intrinsic Grassmann Averages to do online linear, nonlinear and robust PCA in IEEE TPAMI. paper
6/18 Presented our paper on Aggregation of CNNs for boosting performance. paper
5/18 Submitted our paper on Dictionary learning without any explicit sparsity constraint in IEEE TIFT. paper
5/18 Posted our RNN paper on manifold of symmetric positive definite matrices to arxiv. paper | code
5/18 Posted our CNN paper on Riemannian Homogeneous spaces to arxiv. Code will be out shortly. paper
5/18 Our paper on GAN based Autoencoder for fault detection has been accepted in Dx'18 workshop as an Oral paper. paper
4/18 Excited to be serving on the NIPS, BMVC, ACCV, ECCV 2018 PC!
2/18 Our paper on efficient mean estimator on the Stiefel manifold got accepted in Annals of Statistics . paper
1/18 I defended my PhD thesis (Yay! to me). thesis

Research

See my research agenda for details on my current work; below, you find short descriptions of two of my main projects. I am always looking for collaborations. If you're interested in working with me, email me.

In the Geometric Deep Learning project, we are trying to ``generalize'' deep learning algorithms on data resides on non-Euclidean spaces .
Papers DiffCVML18 (CVPR18)

In non-Euclidean statistics project, we developed efficient algorithms to compute statistics on Riemannian manifolds. We made popular sublinear time algorithm to linear/ superlinear and showed weak consistency of our proposed statistical estimators.
Papers Annals of Statistics, ICCV17, CVPR17, CVPR16, ICCV15, MICCAI15

Papers

DMR-CNN: A CNN Tailored for DMR scans with applications to PD classifications
Monami Banerjee*, Rudrasis Chakraborty* and Baba C. Vemuri
IEEE International Symposium on Biomedical Imaging (ISBI) 2019


Statistical Recurrent Models on Manifold valued data
Rudrasis Chakraborty, Chun-Hao Yang*, Xingjian Zhen*, Monami Banerjee, Derek Archer, David Vaillancourt, Vikas Singh and Baba C. Vemuri
NeurIPS 2018


A mixture model for aggregation of multiple pre-trained weak classifiers
Rudrasis Chakraborty, Chun-Hao Yang and Baba C. Vemuri
Diff-CVML (CVPR) 2018
Oral

Generative Adversarial Network based Autoencoder: Application to fault detection problem for closed loop dynamical systems
Rudrasis Chakraborty, Indrasis Chakraborty and Draguna Vrabie
Int. Workshop on Principles of Diagnosis 2018
Oral

Statistics on the (Compact) Stiefel Manifold: Theory and Applications
Rudrasis Chakraborty and Baba C. Vemuri
Annals of Statistics, Vol: 47, No: 1, pp: 415-438, 2019.

A Geometric Framework for Statistical Analysis of Trajectories With Distinct Temporal Spans
Rudrasis Chakraborty, Vikas Singh, Nagesh Adluru, Baba C. Vemuri
ICCV 2017

Sparse Exact PGA on Riemannian Manifolds
Monami Banerjee, Rudrasis Chakraborty and Baba C. Vemuri
ICCV 2017


Statistical Analysis of Longitudinal data and Applications to Neuro-Imaging
Rudrasis Chakraborty and Baba C. Vemuri
ICIP 2017
Oral

Intrinsic Grassmann Averages for Online Linear and Robust Subspace Learning
Rudrasis Chakraborty, Soren Hauberg and Baba C. Vemuri
CVPR 2017

Statistics on the space of trajectories for Longitudinal data analysis
Rudrasis Chakraborty and Baba C. Vemuri
ISBI 2017

An Efficient Exact-PGA Algorithm for Constant Curvature Manifolds
Rudrasis Chakraborty, Dohyung Seo and Baba C. Vemuri
CVPR 2016
Spotlight

A Nonlinear Regression Technique for Manifold Valued Data With Applications to Medical Image Analysis
Monami Banerjee, Rudrasis Chakraborty, Edward Ofori, Michael S. Okun, David E. Vaillancourt and Baba C. Vemuri
CVPR 2016

Recursive Frechet Mean Computation on the Grassmannian and its Applications to Computer Vision
Rudrasis Chakraborty and Baba C. Vemuri
ICCV 2015

Nonlinear Regression on Riemannian Manifolds and Its Applications to Neuro-Image Analysis
Monami Banerjee, Rudrasis Chakraborty, Edward Ofori, David Vaillancourt and Baba C. Vemuri
MICCAI 2015

An efficient recursive estimator of the Fr ́echet mean on hypersphere with applications to Medical Image Analysis
Hesamoddin Salehian, Rudrasis Chakraborty, Edward Ofori, David Vaillancourt and Baba C. Vemuri
MFCA (MICCAI), 2015

An efficient recursive algorithm for atlas construction
Rudrasis Chakraborty, Monami Banerjee, Dohyung Seo, Sara Turner, David Turner, John Forder and Baba C. Vemuri
MFCA (MICCAI), 2015

Students Mentored

Amira Dinari

Teaching

Please see my teaching statement.
F14,15,16 CAP 5416: Computer Vision

Reviewer Services

19 CVPR, ICLR
18 CVPR, ECCV, NIPS, ACCV, BMVC, MICCAI
17 ICCV, CVPR
16 CVPR, BMVC
15 ICCV
16- IEEE TPAMI, IEEE TIP, TSMC-B, TNNLS, IJCV, IEEE TSP, IEEE JBHI