BinaryCoP: Binary Neural Network-based COVID-19 Face-Mask Wear and Positioning Predictor on Edge Devices

Fasfous, Nael; Vemparala, Manoj-Rohit; Frickenstein, Alexander; Frickenstein, Lukas; Badawy, Mohamed A.; Stechele, Walter

doi:10.1109/ipdpsw52791.2021.00024

Cited by 27 publications

(13 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Comparing with INT8, APNN-TC with w2a8 shows lower throughput since we need to compute 16 (=2*8) 1-bit matrices to emulate arbitrary-precision computation, which require more computation than w1a2 with 2 1-bit matrices and w2a2 with 4 1-bit matrices. This also matches the performance on individual kernels (e.g., Figure 5,6,7,8). We also note that APNN-TC can still achieve lower latency on w2a8 than INT8.…”

Section: Apnn Evaluationsupporting

confidence: 83%

“…Arbitrary-precision neural networks have been widely studied to provide diverse tradeoffs between precision and efficiency [6,10,11,24,26,27,36,44,47,49]. While arbitrary-precision may slightly reduce the precision, it shows merit in many practical usages such as smart sensors [23,30,39], mask detection [8], and intelligent agriculture [9]. In these usages, when a certain accuracy bar is surpassed, other essential metrics such as real-time processing and resource consumption are more important.…”

Section: Discussionmentioning

confidence: 99%

“…In these usages, when a certain accuracy bar is surpassed, other essential metrics such as real-time processing and resource consumption are more important. For example, BinaryCoP [8] utilizes low-power binary neural networks to detect facial-mask wear at entrances to corporate buildings and airports. Another example is XpulpNN [9] that uses quantized neural network on energy-efficient IoT devices.…”

Section: Discussionmentioning

confidence: 99%

See 2 more Smart Citations

APNN-TC: Accelerating Arbitrary Precision Neural Networks on Ampere GPU Tensor Cores

Feng¹,

Wang²,

Geng³

et al. 2021

Preprint

View full text Add to dashboard Cite

Over the years, accelerating neural networks with quantization has been widely studied. Unfortunately, prior efforts with diverse precisions (e.g., 1-bit weights and 2-bit activations) are usually restricted by limited precision support on GPUs (e.g., int1 and int4). To break such restrictions, we introduce the first Arbitrary Precision Neural Network framework (APNN-TC) to fully exploit quantization benefits on Ampere GPU Tensor Cores. Specifically, APNN-TC first incorporates a novel emulation algorithm to support arbitrary short bit-width computation with int1 compute primitives and XOR/AND Boolean operations. Second, APNN-TC integrates arbitrary precision layer designs to efficiently map our emulation algorithm to Tensor Cores with novel batching strategies and specialized memory organization. Third, APNN-TC embodies a novel arbitrary precision NN design to minimize memory access across layers and further improve performance. Extensive evaluations show that APNN-TC can achieve significant speedup over CUTLASS kernels and various NN models, such as ResNet and VGG.

show abstract

Section: Apnn Evaluationsupporting

confidence: 83%

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

APNN-TC: Accelerating Arbitrary Precision Neural Networks on Ampere GPU Tensor Cores

Feng¹,

Wang²,

Geng³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Specifically, small, low-cost, and energy efficient computing boards must be used to promote their implementation and enable the use of mobile surveillance while maintaining sufficient computing power and memory to run robust CNN algorithms at a lower latency. Moreover, preserving privacy of pedestrians detected in VSDM systems requires to process data edge devices, without transmitting it to cloud data centers ( Fasfous et al, 2021 ) . In this regard, exploring light-weight CNN algorithms and deploying them on edge and mobile devices can be the great option.…”

Section: Future Directionsmentioning

confidence: 99%

Deep visual social distancing monitoring to combat COVID-19: A comprehensive survey

Himeur

Al‐Maadeed

Almadeed

et al. 2022

Sustainable Cities and Society

View full text Add to dashboard Cite

“…Such algorithms can be used to reduce viral propagation at corporate building entrances, airports, shopping malls, and other indoor locations in the context of the current COVID-19 epidemic. Implementation of BNNs solves a variety of challenges, giving up to 98% accuracy for Masked Face-Net dataset wearing poses [5]. In these dispersed environments capabilities have required low cost battery powered devices with limited memory.…”

Section: Introductionmentioning

confidence: 99%

Masked Face Recognition Using MobileNet V2 with Transfer Learning

Shukla¹,

Tiwari²

2023

Computer Systems Science and Engineering

View full text Add to dashboard Cite

Corona virus (COVID-19) is once in a life time calamity that has resulted in thousands of deaths and security concerns. People are using face masks on a regular basis to protect themselves and to help reduce corona virus transmission. During the on-going coronavirus outbreak, one of the major priorities for researchers is to discover effective solution. As important parts of the face are obscured, face identification and verification becomes exceedingly difficult. The suggested method is a transfer learning using MobileNet V2 based technology that uses deep feature such as feature extraction and deep learning model, to identify the problem of face masked identification. In the first stage, we are applying face mask detector to identify the face mask. Then, the proposed approach is applying to the datasets from Canadian Institute for Advanced Research10 (CIFAR10), Modified National Institute of Standards and Technology Database (MNIST), Real World Masked Face Recognition Database (RMFRD), and Stimulated Masked Face Recognition Database (SMFRD). The proposed model is achieving recognition accuracy 99.82% with proposed dataset. This article employs the four pre-programmed models VGG16, VGG19, ResNet50 and ResNet101. To extract the deep features of faces with VGG16 is achieving 99.30% accuracy, VGG19 is achieving 99.54% accuracy, ResNet50 is achieving 78.70% accuracy and ResNet101 is achieving 98.64% accuracy with own dataset. The comparative analysis shows, that our proposed model performs better result in all four previous existing models. The fundamental contribution of this study is to monitor with face mask and without face mask to decreases the pace of corona virus and to detect persons using wearing face masks.

show abstract

BinaryCoP: Binary Neural Network-based COVID-19 Face-Mask Wear and Positioning Predictor on Edge Devices

Cited by 27 publications

References 15 publications

APNN-TC: Accelerating Arbitrary Precision Neural Networks on Ampere GPU Tensor Cores

APNN-TC: Accelerating Arbitrary Precision Neural Networks on Ampere GPU Tensor Cores

Deep visual social distancing monitoring to combat COVID-19: A comprehensive survey

Masked Face Recognition Using MobileNet V2 with Transfer Learning

Contact Info

Product

Resources

About