[Colloquium] Li/MS Presentation/Jun 23, 2016

[Margaret Jaffey]

This is an announcement of Yun Li's MS Presentation.

------------------------------------------------------------------------------
Date:  Thursday, June 23, 2016

Time:  11:00 AM

Place:  Ryerson 277

M.S. Candidate:  Yun Li

M.S. Paper Title: An Analytical Performance Model for Simultaneous
Localization and Mapping

Abstract:
The emergence of commodity depth sensors provide rich depth data for
mobile devices to achieving Simultaneously Tracking and Mapping (SLAM)
to enable applications like VR or AR; however, the limited computation
resources on mobile devices becomes the bottleneck. This paper
provides an analytical model for SLAM algorithm. We modeled the total
number of instruction counts required for each kernel based on the
input depth data resolution, the volume resolution, and scene
dependent information. To validate the model, we ran SLAM with both
the public available ICL_NUIM dataset and self-collected data with
Kinect V2. We varied the frame rate, depth resolution and volume
resolution to test different configurations. Compared with the
baseline measured by Intel VTune, the model is able to achieve error
rate within 10\%. We also propose Distributed SLAM framework (DSLAM),
which allows different computation kernels of SLAM to be processed by
different devices in a pipelined fashion, so as to lower the
computation requirement of a single device. We think the analytical
model is a possible tool to enable smartly distributing of SLAM
kernels’ computation to available devices.

Yun's advisor is Prof. Andrew Chien

Login to the Computer Science Department website for details:
 https://www.cs.uchicago.edu/phd/ms_announcements#yunli

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Margaret P. Jaffey            margaret at cs.uchicago.edu
Department of Computer Science
Student Support Rep (Ry 156)               (773) 702-6011
The University of Chicago      http://www.cs.uchicago.edu
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=