[Colloquium] 2/6 Thesis Defense: Payman Yadollahpour, TTIC

[Mary Marre via Colloquium]

When:    Monday, February 6th at 1:00 pm

Where:   TTIC, 6045 S Kenwood Avenue, 5th Floor, Room 526

Who:      Payman Yadollahpour, TTIC


Title: Exploring and Exploiting Diversity for Image Segmentation

Abstract:
Semantic image segmentation is an important computer vision task that is
difficult
because it consists of both recognition and segmentation. The task is often
cast
as a structured output problem on an exponentially large output-space,
which is
typically modeled by a discrete probabilistic model. The best segmentation
is found by inferring the Maximum a-Posteriori (MAP) solution over the
output
distribution defined by the model. Due to limitations in optimization, the
model cannot be
arbitrarily complex. This leads to a trade-off: devise a more accurate
model that
incorporates rich high-order interactions between image elements at the
cost of
inaccurate and possibly intractable optimization OR leverage a tractable
model which
produces less accurate MAP solutions but may contain high quality solutions
as other
modes of its output distribution.

This thesis investigates the latter and presents a two
stage approach to semantic segmentation akin to cascade models and proposal
generation works. In the first stage a tractable probabilistic
model outputs a set of high probability segmentations from the underlying
distribution that are not just minor perturbations of each other.
Critically the
output of this stage is a diverse set of plausible solutions and not just a
single one. The first-stage reduces the exponential space of solutions to
just a
handful of segmentations. In the second stage, a discriminatively trained
re-ranking
model selects the best segmentation from this set. The re-ranking stage can
use much more complex features than what could be tractably used in the
probabilistic model, allowing a better exploration of the solution space
than possible
by simply producing the most probable solution from the probabilistic
model. The
formulation of the first-stage is agnostic to the underlying model and
optimization algorithm, which makes it applicable to a wide-range of models
and
inference methods.

Evaluation of the approach on a number of semantic image segmentation
benchmark datasets highlight its superiority over inferring the MAP
solution.


Thesis Advisor: Gregory Shakhnarovich, greg at ttic.edu



Mary C. Marre
Administrative Assistant
*Toyota Technological Institute*
*6045 S. Kenwood Avenue*
*Room 504*
*Chicago, IL  60637*
*p:(773) 834-1757*
*f: (773) 357-6970*
*mmarre at ttic.edu <mmarre at ttic.edu>*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20170130/77e883cb/attachment.html>