Namdar Homayounfar

namdar@{cs, stat}.toronto.edu

I'm a PhD candidate in the department of Statistical Sciences at University of Toronto and am also a member of the Machine Learning group within the department of Computer Science. I'm advised by Prof. Raquel Urtasun and Prof. Sanja Fidler. Previously, I obtained a MSc in Statistics from University of Toronto and BSc in Probability and Statistics from McGill University. Please see my CV for more info.

Research Interests

My main area of interest is Statistical Machine Learning and Computer Vision with applications to sports analytics and autonomous driving.

Publications & Submissions

2017:

- Sports Field Localization Via Deep Structured Models

N. Homayounfar, S. Fidler, R. Urtasun
To appear in Conference on Computer Vision and Pattern Recognition (CVPR), Hawaii, US, July 2017
Description In this work, we propose a novel way of efficiently localizing a sports field from a single broadcast image image of the game. Related work in this area relies on manually annotating a few key frames and extending the localization to similar images, or installing fixed specialized cameras in the stadium from which the layout of the field can be obtained. In contrast, we formulate this problem as a branch and bound inference in a Markov random field where an energy function is defined in terms of semantic cues such as the field surface, lines and circles obtained from a deep semantic segmentation network. Moreover, our approach is fully automatic and depends only on a single image from the broadcast video of the game. We demonstrate the effectiveness of our method by applying it to soccer and hockey.
[paper pdf] [supp pdf (139 mb)] [soccer data (57 mb)] [Code to be released soon]

2015:

- Periodic Solutions of a Singularly Perturbed Delay Differential Equation with Two State-Dependent Delays

A. R. Humphries, D. A. Bernucci, R. C. Calleja, N. Homayounfar, M. Snarski
Journal of Dynamics and Differential Equations
Description Delay Differential Equations (DDEs) are a certain class of differential equations in which the rate of change of a system depends not only on its current state, which is the case for ordinary differential equations, but also on its state in the past. These equations arise naturally when modelling various biological and physical phenomena and provide a glimpse of the often complex and chaotic dynamics of such systems. In this paper, we study the periodic solutions of a certain class of DDEs.
[PDF] [Journal]

2013:

- MCMC Clustering and Its Convergence Issues

N. Homayounfar, M. Asgharian, V. Partovi Nia
Contributed Poster in JSM
Description Bayesian clustering using MCMC sampling is a popular approach. When a Markov chain Monte Carlo method is applied, the Markov chain samples are used to approximate the posterior after the chain is converged. When the data grouping is the concern, the convergence must be checked over the allocation space. The convergence of a Markov chain is verified, often using a trace plot, or using other common quantitative criteria mostly designed for a continuous state space. However, data allocation is a very large unordered discrete space and therefore the common convergence criteria is nontrivial to apply. We monitor the convergence of a clustering chain by a convergence criterion devised for the data allocation space.
[Poster PDF] [Conference Entry]

Curriculum Vitae

Please download my CV for more information.