Search

Fatih S Porikli

from San Diego, CA
Age ~52

Fatih Porikli Phones & Addresses

  • San Diego, CA
  • Carlsbad, CA
  • 29 Sylvanus Wood Ln, Woburn, MA 01801
  • Watertown, MA
  • North Plainfield, NJ
  • Cambridge, MA
  • Brooklyn, NY

Work

Company: Mitsubishi electric research labs (merl) Mar 2000 Position: Distinguished research scientist (since 2011 march), technical manager

Education

Degree: PhD School / High School: Polytechnic University 1996 to 2002 Specialities: Electrical and Computer Engineering, Computer Science

Skills

Computer Vision • Machine Learning • Signal Processing • Algorithms • Pattern Recognition • Image Processing • Data Mining • Artificial Intelligence • R&D • Robotics • Video Processing • Computer Science • Software Engineering • C++ • Simulations • Programming • Algorithm Design • Mathematical Modeling • Technology Management • Optimization • Digital Signal Processors • High Performance Computing • Machine Vision • Embedded Systems • Numerical Analysis • Matlab • Proposal Writing • Distributed Systems • Sensors • Digital Signal Processing • Opengl • Robot Framework • Software Development • Research and Development • High Performance Computing

Interests

Ieee Cvpr 2012 • Nsf Panelist Judge • Ieee Avss 2012 • Usa Liaison • 2003 To Present • Iapr Icpr 2010 • Editor • 2011 To Present • Spie Real Time Imaging • Special Tracks Chair • 3 Other Guest Editors • Ieee Intelligent Vehicles 2009 • 2004 To Present • Advisory Board • Associate Editor • Journal of Machine Vision Applications • Industrial Chair • 2011 • Ieee Iccv 2011 • Ieee Avss 2010 • Area Chair • Ieee Icme 2007 • Ieee Cvpr 2009 • See 12 • Vcip 2004 • Corporate Relations Chair • Ieee Signal Processing Magazine • Siam Imaging Sciences • Program Chair • Springer • Technical Program Chair • Track Chair • See Less • 2010 • 2006 To Present • Isvc 2009 • 2008 • Co Editor

Industries

Information Technology And Services

Resumes

Resumes

Fatih Porikli Photo 1

Vice President Of Computer Vision Lab

View page
Location:
3498 Sitio Borde, Carlsbad, CA 92009
Industry:
Information Technology And Services
Work:
Mitsubishi Electric Research Labs (MERL) since Mar 2000
Distinguished Research Scientist (since 2011 March), Technical Manager

Hughes Research Laboratories, Malibu, CA May 1999 - Mar 2000
Computer Vision Expert (Contractor)

AT&T Bell Labs, Holmdel, NJ Sep 1997 - Aug 1998
Advanced Developer (Contractor)
Education:
Polytechnic University 1996 - 2002
PhD, Electrical and Computer Engineering, Computer Science
Polytechnic University, Brooklyn, NY Sep 1994 - Jul 1996
MS, Electrical and Computer Engineering, Mathematics
Bilkent University Sep 1988 - Jan 1992
BS, Electrical Engineering
Skills:
Computer Vision
Machine Learning
Signal Processing
Algorithms
Pattern Recognition
Image Processing
Data Mining
Artificial Intelligence
R&D
Robotics
Video Processing
Computer Science
Software Engineering
C++
Simulations
Programming
Algorithm Design
Mathematical Modeling
Technology Management
Optimization
Digital Signal Processors
High Performance Computing
Machine Vision
Embedded Systems
Numerical Analysis
Matlab
Proposal Writing
Distributed Systems
Sensors
Digital Signal Processing
Opengl
Robot Framework
Software Development
Research and Development
High Performance Computing
Interests:
Ieee Cvpr 2012
Nsf Panelist Judge
Ieee Avss 2012
Usa Liaison
2003 To Present
Iapr Icpr 2010
Editor
2011 To Present
Spie Real Time Imaging
Special Tracks Chair
3 Other Guest Editors
Ieee Intelligent Vehicles 2009
2004 To Present
Advisory Board
Associate Editor
Journal of Machine Vision Applications
Industrial Chair
2011
Ieee Iccv 2011
Ieee Avss 2010
Area Chair
Ieee Icme 2007
Ieee Cvpr 2009
See 12
Vcip 2004
Corporate Relations Chair
Ieee Signal Processing Magazine
Siam Imaging Sciences
Program Chair
Springer
Technical Program Chair
Track Chair
See Less
2010
2006 To Present
Isvc 2009
2008
Co Editor

Publications

Us Patents

Method For Determining Compactness Ratios Of Multiple Data And Signal Sets

View page
US Patent:
6885765, Apr 26, 2005
Filed:
Apr 4, 2001
Appl. No.:
09/826048
Inventors:
Fatih M. Porikli - North Plainfield NJ, US
Assignee:
Mitsubishi Electric Research Laboratories, Inc. - Cambridge MA
International Classification:
G06K009/62
G06K009/00
US Classification:
382159, 382160, 382165, 382170
Abstract:
A method determines a compactness ratio for data sets by first measuring a combined compactness value for a union of the data sets, and second, by measuring an individual compactness value for each one of the data set. The combined compactness value is then divided by a sum of the individual compactness values to determine the compactness ration of the data sets.

Identifying Moving Objects In A Video Using Volume Growing And Change Detection Masks

View page
US Patent:
6904159, Jun 7, 2005
Filed:
Dec 20, 2001
Appl. No.:
10/026984
Inventors:
Fatih M. Porikli - North Plainfield NJ, US
Assignee:
Mitsubishi Electric Research Laboratories, Inc. - Cambridge MA
International Classification:
G06K009/00
US Classification:
382103
Abstract:
The method combines volume growing with change detection. After an input video is filtered to remove noise, a spatio-temporal data structure is formed from the video frames, and markers are selected. From the markers, volumes are grown using a color similarity based centroid linkage method. Change detection masks are then extracted from adjacent frames in the video using local color features. The change detection masks are intersected with each volume, to determine the number of changed pixels only in portions of the masks that lie within that volume. If the number of changed pixels in the intersection exceeds a threshold, then the volume is identified as a moving object.

Method And Apparatus For Decoding Video Bitstreams To Reduced Spatial Resolutions

View page
US Patent:
7006572, Feb 28, 2006
Filed:
May 9, 2002
Appl. No.:
10/141742
Inventors:
Fatih M. Porikli - North Plainfield NJ, US
Huifang Sun - Murray Hill NJ, US
Assignee:
Mitsubishi Electric Research Labs, Inc. - Cambridge MA
International Classification:
H04N 7/12
US Classification:
37524016
Abstract:
A method decodes a video encoded as a compressed bitstream including frames constructed as blocks. The bitstream is variable length decoded and inverse quantized to determine motion vectors and DCT coefficients of the blocks of the bitstream. The DCT coefficients are then inverse discrete cosine transformed. A low-resolution block is generated for each reconstructed block, and the reconstructed blocks are motion compensated with the low-resolution blocks to decode the video.

Method And System For Minimizing Error In Bandwidth Allocation With An Optimal Number Of Renegotiations

View page
US Patent:
7027403, Apr 11, 2006
Filed:
May 22, 2001
Appl. No.:
09/862900
Inventors:
Fatih M. Porikli - North Plainfield NJ, US
Zefer Sahinoglu - Clifton NJ, US
Assignee:
Mitsubishi Electric Research Laboratories, Inc. - Cambridge MA
International Classification:
H04J 3/14
H04L 12/26
US Classification:
370238, 370252, 37039541, 709241
Abstract:
A method dynamically allocates and renegotiates bandwidth to traffic having a variable data rate in a network. A current data rate and current allocated bandwidth is measured. A future data rate for the traffic is predicted. A cost function is then minimized based on the current data rate, the current allocated bandwidth, and the future data rate to determine a future bandwidth to be allocated for the traffic so that a cost of the renegotiation is minimized over time.

Image Simplification Using A Robust Reconstruction Filter

View page
US Patent:
7103229, Sep 5, 2006
Filed:
Nov 19, 2001
Appl. No.:
10/012915
Inventors:
Fatih M. Porikli - North Plainfield NJ, US
Assignee:
Mitsubishi Electric Research Laboratories, Inc. - Cambridge MA
International Classification:
G06K 9/40
G06K 9/36
US Classification:
382275, 382276
Abstract:
A method simplifies a data structure representing physical measurements of a real-world phenomena, such as an image of a scene of an object. Input data are first acquired, sensed, or measured. If the data are acquired with a camera, then measurement errors for data points in the data structure are not normally distributed. Therefore, an error term for each data point is constructed according to a Lorentzian estimator. The error term can be determined by taking a difference between the value of the data point, and the value at the data point when a model is fitted to the data structure. The error term is then minimized using a downhill simplex minimization process. Finally, each data point is adjusted by the minimized error term to produce a simplified data structure of the real-world phenomena.

Method For Segmenting 3D Objects From Compressed Videos

View page
US Patent:
7142602, Nov 28, 2006
Filed:
May 21, 2003
Appl. No.:
10/442417
Inventors:
Fatih M. Porikli - Watertown MA, US
Huifang Sun - Cambridge MA, US
Ajay Divakaran - Burlington MA, US
Assignee:
Mitsubishi Electric Research Laboratories, Inc. - Cambridge MA
International Classification:
H04B 1/66
US Classification:
37524016
Abstract:
A method segments a video into objects, without user assistance. An MPEG compressed video is converted to a structure called a pseudo spatial/temporal data using DCT coefficients and motion vectors. The compressed video is first parsed and the pseudo spatial/temporal data are formed. Seeds macro-blocks are identified using, e. g. , the DCT coefficients and changes in the motion vector of macro-blocks. A video volume is “grown” around each seed macro-block using the DCT coefficients and motion distance criteria. Self-descriptors are assigned to the volume, and mutual descriptors are assigned to pairs of similar volumes. These descriptors capture motion and spatial information of the volumes. Similarity scores are determined for each possible pair-wise combination of volumes. The pair of volumes that gives the largest score is combined iteratively.

Method And System For Assigning Circuits To A New Service Request In A Communications Network

View page
US Patent:
7209438, Apr 24, 2007
Filed:
May 22, 2001
Appl. No.:
09/862899
Inventors:
Zafer Sahinoglu - Clifton NJ, US
Fatih M. Porikli - North Plainfield NJ, US
Fernando M. Matsubara - Gillette NJ, US
Johnas Cukier - Chester NY, US
Assignee:
Mitsubishi Electric Research Laboratories, Inc. - Cambridge MA
International Classification:
G08C 15/00
H04J 3/14
H04L 12/26
US Classification:
370230
Abstract:
A system and method selects a circuit to service an application request to transmit data over a network. The network includes one or more low and high bandwidth circuits. An average utilization is measured for each circuit in a circuit analyzer should the application request be assigned to the circuit. Then, application request is assigned to the high bandwidth circuit if the average utilization is less than a predetermined threshold, assigning to the low bandwidth circuit if the average utilization is less than one, where one is a full utilization, and declined otherwise. The threshold is one minus a guard bandwidth of the high bandwidth channel.

Adaptive Background Image Updating

View page
US Patent:
7224735, May 29, 2007
Filed:
May 21, 2003
Appl. No.:
10/442752
Inventors:
Fatih M. Porikli - Watertown MA, US
Oncel Tuzel - Piscataway NJ, US
Dirk Brinkman - Cambridge MA, US
Assignee:
Mitsubishi Electronic Research Laboratories, Inc. - Cambridge MA
International Classification:
H04N 7/18
US Classification:
37524025, 37534026
Abstract:
A method compares a background image to input images to determine a similarity scores λ for each input image. Then, the background image is updated only if the similarity score for a particular image is less than a predetermined threshold. Presumably, any pixel whose color does not change is part of a static background, and any pixel that does change is part of a moving object. The similarity score controls when input images are scored and the manner the background image is updated.
Fatih S Porikli from San Diego, CA, age ~52 Get Report