Face detection algorithm based on improved Retinaface

Chen, Yanfei; Wu, ZeCheng; Peng, Hongsheng; Zhou, Chao; Yan, Zhangchen; Huang, Yuliang

doi:10.1109/cac53003.2021.9728659

Cited by 5 publications

(2 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One-stage detector with some typical models such as: SSD [19], Yolo [20], RetinaNet [21,22], [23]. Called one-stage because in the design of the model, there is absolutely no extraction of feature regions (areas that can contain objects) like RPN [24] or Faster-RCNN [25].…”

Section: Source: Compiled By the Authorsmentioning

confidence: 99%

A survey on deep learning based face detection

Tran¹,

Nguyen²,

Tran³

2023

AAIT

View full text Add to dashboard Cite

The article has focused on surveying face detection models based on deep learning, specifically examining different one-stage models in order to determine how to choose the appropriate face detection model as well as propose a direction to enhance our face detection model to match the actual requirements of computer vision application systems related to the face. The face detection models that were conducted survey include single shot detector, multi-task cascaded convolution neural networks, RetinaNet, YuNet on the Wider Face dataset. Tasks during the survey are structural investigation of chosen models, conducting experimental surveys to evaluate the accuracy and performance of these models. To evaluate and provide criteria for choosing face detection suitable for the requirements, two indicators are used, average precision to evaluate accuracy and frames-per-second to evaluate performance. Experiential results were analyzed and used for making conclusions and suggestions for future work. For our real-time applications on face-related camera systems, such as driver monitoring system, supermarket security system (shoplifting warning, disorderly warning), attendance system, often require fast processing, but still ensures accuracy. The models currently applied in our system such as Yolos, Single Shot Detector, MobileNetv1 guarantee real-time processing, but most of these models have difficulty in detecting small faces in the frame and cases containing contexts, which are easily mistaken for a human face. Meanwhile, the RetinaNet_ResNet50 model brings the highest accuracy, especially to ensure the detection of small faces in the frame, but the processing time is larger. Therefore, through this survey, we propose an enhancement direction of the face detection model based on the RetinaNet structure with the goal of ensuring accuracy and reducing processing time.

show abstract

Section: Source: Compiled By the Authorsmentioning

confidence: 99%

A survey on deep learning based face detection

Tran¹,

Nguyen²,

Tran³

2023

AAIT

View full text Add to dashboard Cite

show abstract

“…However, this approach faces the challenge of significant underlying blockchain communication latency. The literature [19] employs a hierarchical blockchain approach with main and sub-chains to address the issue of secure identity verification for drone clusters across trust domains and diverse network environments. However, this method's efficiency for authentication is affected to some extent as the number of authentication nodes increases, leading to an increase in data transmission volume.…”

Section: Introductionmentioning

confidence: 99%

A Hierarchical Blockchain-Based Trust Measurement Method for Drone Cluster Nodes

Zuo,

Cao,

et al. 2023

Drones

View full text Add to dashboard Cite

In response to the challenge of low accuracy in node trust evaluation due to the high dynamics of entry and exit of drone cluster nodes, we propose a hierarchical blockchain-based trust measurement method for drone cluster nodes. This method overcomes the difficulties related to trust inheritance for dynamic nodes, trust re-evaluation of dynamic clusters, and integrated trust calculation for drone nodes. By utilizing a multi-layer unmanned cluster blockchain for trusted historical data storage and verification, we achieve scalability in measuring intermittent trust across time intervals, ultimately improving the accuracy of trust measurement for drone cluster nodes. We design a resource-constrained multi-layer unmanned cluster blockchain architecture, optimize the computing power balance within the cluster, and establish a collaborative blockchain mechanism. Additionally, we construct a dynamic evaluation method for trust in drone nodes based on task perception, integrating and calculating the comprehensive trust of drone nodes. This approach addresses trusted sharing and circulation of task data and resolves the non-inheritability of historical data. Experimental simulations conducted using NS3 and MATLAB demonstrate the superior performance of our trust value measurement method for unmanned aerial vehicle cluster nodes in terms of accurate malicious node detection, resilience to trust value fluctuations, and low resource delay retention.

show abstract

An Improved Attendance Monitoring System through Facial Recognition using RetinaFace Algorithm

Dhanalakshmi,

M G

et al. 2024

2024 3rd International Conference for Innovation in Technology (INOCON)

View full text Add to dashboard Cite

Face detection algorithm based on improved Retinaface

Cited by 5 publications

References 16 publications

A survey on deep learning based face detection

A survey on deep learning based face detection

A Hierarchical Blockchain-Based Trust Measurement Method for Drone Cluster Nodes

An Improved Attendance Monitoring System through Facial Recognition using RetinaFace Algorithm

Contact Info

Product

Resources

About