Prototype Development of Face and Speaker Recognitions based on Edge Computing

Pan, Yuxuan; Peng, Xiaowen; Lin, Xuexian; Xia, Minghua

doi:10.1109/isncc49221.2020.9297243

Cited by 1 publication

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multi-Task Cascaded Convolutional Neural Network (MTCNN) is a convolution network developed by the Chinese Academy of Sciences [82]. Its architecture consists of three stages of convolutional networks that predict the face landmarks to detect the faces in an image [2]. The first stage is a proposal network that will predict potential face positions and bound rectangular or elliptical boxes around the predicted locations.…”

Section: ) Mtcnnmentioning

confidence: 99%

“…MTCNN which has been successfully used for face detection in ( [2], [84], [85], [86]), features an advantage over FaceNet is that it can simultaneously detect more than one face in an image and feed them to a recognition system. On the other hand, FaceNet can detect only one face at a time [2]. However, one of its major disadvantages is that it can't cope with face rotations [87].…”

Section: ) Mtcnnmentioning

confidence: 99%

“…However, the suitability of the algorithm to run on IoT/edge devices was not investigated and was recommended as major future goal by the authors. In [2] a face recognition network cascaded to a face detection network is presented to run on an edge device (x86 platform). The detection is based on the MTCNN algorithm, and the FaceNet algorithm is used to recognize the detected faces.…”

Section: ) Facenetmentioning

confidence: 99%

“…T HE use of biometric features, such as the face, fingerprint, voice, hand geometry, and retina eye, which are based on human's unique biological, physical and behavioral characteristics has been widely used for identification and recognition tasks [1]. These features have the advantage over other computer security tools, such as passwords, confidential codes, or handheld tokens to be more secure [2]. Among all the aforementioned biometric features, human faces remain one of the most effective features used for human identification since the associated systems can be contactless and can handle simultaneously several humans.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Hardware Accelerators for Real-Time Face Recognition: A Survey

2022

View full text Add to dashboard Cite

Real-time face recognition has been of great interest in the last decade due to its wide and variant critical applications which include biometrics, security in public places, and identification in login systems. This has encouraged researchers to design fast and accurate embedded and portable systems that are capable to detect and recognize a large number of faces at almost video frame rate. Due to the increasing volume of reference faces, traditional general purpose computing engines such as the ones based on Intel's Pentium processors have shown not to be adequate and various dedicated hardware accelerators based on either Graphical Processing Unit (GPU), Field Programmable Gate Arrays (FPGA), Application Specific Integrated Circuit (ASIC), or even multi-core Central Processing Units (CPU) have emerged. Earlier published review papers on face detection/recognition have discussed face detection and face recognition algorithms enhancement that improve the detection accuracy. Nevertheless, none of them has reviewed the hardware accelerators used for this application. Accordingly, this paper aims to provide a comprehensive review of the most recent face recognition algorithms and associated embedded hardware systems targeting real-time performance. A detailed comparison between neural network and non-neural network-based algorithms in terms of accuracy and processing time is provided. Discussions on their suitability to be implemented into parallel hardware architectures such as Single Instruction Multiple Thread (SIMT) or Single Instruction Multiple Data (SIMD) is also discussed.

show abstract

Section: ) Mtcnnmentioning

confidence: 99%