Fast and Robust Face Detection on a Parallel Optimized Architecture Implemented on FPGA

Farrugia, Nicolas; Mamalet, Franck; Roux, Sébastiên; Yang, Fan; Paindavoine, Michel

doi:10.1109/tcsvt.2009.2014013

Cited by 30 publications

(14 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, since the features of each stage are different in number and they cannot just be the multiple of classifier numbers, the structure is wasteful in some sense in terms of hardware. Considering a majority of sub-windows are rejected at the first several stages, it is reasonable to propose a partially parallel architecture for classification [13], [15]. Frequently executed tasks could be executed in a parallel processing module and those less frequently could be executed in a sequential processing model.…”

Section: Classify Processing Modulementioning

confidence: 99%

A low cost architecture for high performance face detection

Zhou

Zeng

2015

Microprocessors and Microsystems

View full text Add to dashboard Cite

Section: Classify Processing Modulementioning

confidence: 99%

A low cost architecture for high performance face detection

Zhou

Zeng

2015

Microprocessors and Microsystems

View full text Add to dashboard Cite

“…For example, FPGAs can help accelerate the ViolaJones algorithm by allowing multiple features to be computed in parallel [1,16]. Farrugia et al [5] also used FPGAs to accelerate the face detection process, but they relied on pipelining convolution operations to achieve a speed boost using a neural network based face detector. Hefenbrock et al [6] investigated using Graphical Processing Units (GPUs) to accelerate the Viola-Jones algorithm by using the GPU to scan multiple windows for faces in parallel.…”

Section: Background Workmentioning

confidence: 99%

“…In particular, as the number of faces being detected increases or the resolution of the image being processed goes higher, the performance of face detection applications degrades drastically. Various parallel face detection systems have been proposed that can run on dedicated hardware and perform face detection in real-time for high resolution images [1,5,6], but development of face detection systems that utilize the potentials of multiple cores on consumer level processors has largely been missing. Considering that multi-core processors are now standard in the latest PCs and mobile devices, we believe that rethinking the face detection pipeline to make effective use of these cores is a fruitful area for research.…”

Section: Introductionmentioning

confidence: 99%

Parallelizing a Face Detection and Tracking System for Multi-Core Processors

Ranjan¹,

Malik²

2012

2012 Ninth Conference on Computer and Robot Vision

View full text Add to dashboard Cite

This paper describes how to accelerate a real-world face detection and tracking system by taking advantage of the multiple processing cores that are present in most modern CPUs. This work makes three key contributions. The first is the presentation of a highly optimized serial face detection and tracking algorithm that uses motion estimation and local search windows to achieve fast processing rates. The second is redefining the face detection process based on a set of independent face scales that can be processed in parallel on separate CPU cores while also achieving a target processing rate. The third contribution is demonstrating how multiple cores can be used to accelerate the face tracking process which provides significant speed boosts when tracking a large number of faces simultaneously. Used in a real-world application, the parallel face detector and tracker yields a 50-70% speed boost over the serial version when tested on a commodity multi-core CPU.

show abstract

“…The processing power on these devices is limited for traditional video processing tasks like face recognition. Although advanced hardware design and algorithm optimization could be helpful to certain extent [32], most of existing video-based face recognition approaches require high computation, which prevents them for wide applications in low-performance systems. Therefore, there is a great need to investigate low-cost algorithms for face recognition in video.…”

Section: Challenges and Future Directionsmentioning

confidence: 99%

Face Recognition and Retrieval in Video

Shan¹

2010

Video Search and Mining

View full text Add to dashboard Cite

Abstract. Automatic face recognition has long been established as one of the most active research areas in computer vision. Face recognition in unconstrained environments remains challenging for most practical applications. In contrast to traditional still-image based approaches, recently the research focus has shifted towards videobased approaches. Video data provides rich and redundant information, which can be exploited to resolve the inherent ambiguities of image-based recognition like sensitivity to low resolution, pose variations and occlusion, leading to more accurate and robust recognition. Face recognition has also been considered in the content-based video retrieval setup, for example, character-based video search. In this chapter, we review existing research on face recognition and retrieval in video. The relevant techniques are comprehensively surveyed and discussed.

show abstract

Fast and Robust Face Detection on a Parallel Optimized Architecture Implemented on FPGA

Cited by 30 publications

References 19 publications

A low cost architecture for high performance face detection

A low cost architecture for high performance face detection

Parallelizing a Face Detection and Tracking System for Multi-Core Processors

Face Recognition and Retrieval in Video

Contact Info

Product

Resources

About