Record Details

Real Time Face Recognition on GPU using OPENCL

Electronic Theses of Indian Institute of Science

View Archive Info
 
 
Field Value
 
Title Real Time Face Recognition on GPU using OPENCL
 
Creator Naik, Narmada
 
Subject Real Time Face Recognition
Face Recognition
GPU
Local Binary Pattern (LBP)
Video Based Face Recognition
Tracking after Recognition
Local Ternary Pattern (LTP)
Enhanced Local Ternary Patterns (ELTP)
Face Identification
Face Identification from Depth (RGBD)
Kinect Sensor
OpenCL Memory Model
Electrical Engineering
 
Description Face recognition finds various applications in surveillance, Law enforcement etc. These applications require fast image processing in real time. Modern GPUs have evolved fully programmable parallel stream processors. The problem of face recognition in real time system is benefited by parallelism. With the aim of fulfilling both speed and accuracy criteria we present a GPU accelerated Face Recognition system. OpenCL is a heterogeneous computing language that allows extracting parallelism on diļ¬€erent platforms like DSP processors, FPGAs, GPUs. The proposed kernel on GPU exploits coarse grain parallelism for Local Binary Pattern (LBP) histogram computation and ELTP (Enhanced Local Ternary Pattern) feature extraction. The proposed optimization techniques on local memory, work group size and work group dimension enhances the computation of face recognition on GPU further. As a result, we have achieved a speed up of 30 times to 300 times for 124*124 to 2048*2048 image sizes for LBP and ELTP feature extraction compared to CPU. We also present a robust real time face recognition and tracking on GPU using fusion of RGB and Depth images taken from Kinect sensor. The proposed segmentation after detection algorithm enhances the performances of recognition using LBP.
 
Contributor Rathna, G N
 
Date 2018-05-23T15:47:06Z
2018-05-23T15:47:06Z
2018-05-23
2017
 
Type Thesis
 
Identifier http://etd.iisc.ernet.in/2005/3596
http://etd.iisc.ernet.in/abstracts/4465/G28212-Abs.pdf
 
Language en_US
 
Relation G28212