Face


email:
balakg@mit.edu

github:
github.com/balakg

address:
32 Vassar St.
Cambridge, MA
02139

Guha Balakrishnan

I am currently a postdoc in Bill Freeman’s group at MIT, with research interests in computer vision, graphics and machine learning. I completed my Master’s (2013) and PhD (2018) degrees in the EECS dept. at MIT, advised by Professors John Guttag and Frédo Durand. My PhD thesis was titled Analyzing and Synthesizing Deformations in Image Datasets. I explored how modeling deformations can be useful for various applications such as image warping/synthesis, image registration, and action alignment in videos.

I also have an interest in applications of computer vision to healthcare. My Master's work focused on vitals signs measurement from video recordings.

Prior to MIT, I completed my bachelor of science degrees in Computer Science and Computer Engineering at the University of Michigan, Ann Arbor. There, I worked on medical machine learning research with Professor Zeeshan Syed.

Recent News:

I am excited to present the following papers at CVPR 2018:

An Unsupervised Learning Model for Deformable Medical Image Registration.
Guha Balakrishnan, Amy Zhao, Mert R. Sabuncu, John Guttag, Adrian V. Dalca.
paper, code

Synthesizing Images of Humans in Unseen Poses. (oral)
Guha Balakrishnan, Amy Zhao, Adrian V. Dalca, Frédo Durand, John Guttag.
paper, code

Other Work:

Unsupervised Learning for Fast Probabilistic Diffeomorphic Registration.
Adrian V. Dalca, G. Balakrishnan, J. Guttag, and Mert R. Sabuncu.
MICCAI 2018.

Video diff: highlighting differences between similar actions in videos.
Guha Balakrishnan,  Frédo Durand, John Guttag.
Siggraph Asia 2015.
project page

Detecting Pulse from Head Motions in Video.
Guha Balakrishnan,  Frédo Durand, John Guttag.
CVPR 2013.
project page

A Video-Based Method for Objectively Rating Ataxia.
Ronnachai Jaroensri, Amy Zhao, Guha Balakrishnan, Derek Lo, Jeremy Schmahmann, John Guttag, Fredo Durand
MLHC 2017.
paper

Scalable personalization of long-term physiological monitoring: Active learning methodologies for epileptic seizure onset detection.
Guha Balakrishnan, Zeeshan Syed.
AISTATS 2012.
paper