Face Detection on a parallel platform using CUDA technology
| Author(s) | : | Patel Raksha R, Ms. Isha K. Vajani |
| Institution | : | M.Tech Student, Computer Department CGPIT, UTU, Bardoli, Gujarat, India |
| Published In | : | Vol. 2, Issue 5 — May 2015 |
| Page No. | : | 1409-1417 |
| Domain | : | Engineering |
| Type | : | Research Paper |
| ISSN (Online) | : | 2348-4470 |
| ISSN (Print) | : | 2348-6406 |
Face Detection finds an application in various fields like live video surveillance, bio-metrics, augmented reality,etc. However, CPU single thread implementation of face detection consumes lot of time, and despite various optimizationtechniques, it performs poorly at real time. Our design model combines conventional programming using CPU with multicore programming using NVIDIA CUDA for fast face detection. The face detection algorithm can be paralyzed to implementon the GPU using CUDA technology. Viola-Jones face algorithm when exploited in parallel, it gives optimized output. Thespeed up factor for various algorithms has been calculated. The algorithms involving rigorous computations can performbetter on parallel platforms when broken up into parallel and executed concurrently.
Patel Raksha R, Ms. Isha K. Vajani, “Face Detection on a parallel platform using CUDA technology”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 2, Issue 5, pp. 1409-1417, May 2015.
