High performance architecture for object detection in streamed videos

Author

Zemcik, Pavel ; Juranek, Roman ; Musil, Petr ; Musil, Martin ; Hradis, Michal

Author_Institution

Dept. of Comput. Graphics & Multimedia, Brno Univ. of Technol., Brno, Czech Republic

fYear

2013

fDate

2-4 Sept. 2013

Firstpage

1

Lastpage

4

Abstract

In this paper, we introduce a novel architecture of an engine for high performance multi-scale detection of objects in videos based on WaldBoost training algorithm. The key properties of the architecture include processing of streamed data and low resource consumption. We implemented the engine in FPGA and we show that it can process 640×480 pixel video streams at over 160 fps without the need of external memory. We evaluate the design on the face detection task, compare it to state of the art designs, and discuss its features and limitations.

Keywords

face recognition; field programmable gate arrays; object detection; video streaming; FPGA; WaldBoost training algorithm; face detection task; high performance architecture; high performance multiscale detection; object detection; pixel video streams; resource consumption; streamed data processing; streamed videos; Computer architecture; Detectors; Engines; Feature extraction; Field programmable gate arrays; Object detection; Videos;

fLanguage

English

Publisher

ieee

Conference_Titel

Field Programmable Logic and Applications (FPL), 2013 23rd International Conference on

Conference_Location

Porto

Type

conf

DOI

10.1109/FPL.2013.6645559

Filename

6645559