YOLO5Face: Why Reinventing a Face Detector

Qi, Delong; Tan, Weijun; Yao, Qi; Liu, Jingfeng

doi:10.1007/978-3-031-25072-9_15

Cited by 93 publications

(34 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We also included three well performing and known algorithms, MTCNN, 34 RetinaFace 35 and yolov5Face. 36 We applied the algorithms in the literature with moderate network sizes that could fit in the SoC. In that manner, MTCNN is used as is without any size modifications, while RetinaFace and yolov5Face are applied in (640x640) and (416x416) resolutions respectively.…”

Section: Resultsmentioning

confidence: 99%

Simultaneous license plate recognition and face detection application at the edge

Alim

Kavakli

Okcu

et al. 2023

AI and Optical Data Sciences IV

View full text Add to dashboard Cite

Face recognition (FR) and license plate recognition (LPR) are very crucial algorithms for identification of humans and vehicles in several applications such as surveillance, traffic and access-control. The advances in small singleboard computers with high parallel processing power capabilities and the use of low-power Neural Processing Units (NPU) inside embedded System on Chips (SoC), enable real-time face detection (FD) and LPR at the edge. On the other hand, it is still a challenge to run multiple algorithms concurrently with high accuracy and prompt execution (high frame rates) that requires a very efficient software/video analytics algorithm development. Both FR and LPR algorithms need two-stage processing that involve detection and recognition. In this study, we propose a method that enables simultaneous face detection associated with landmark and quality information and LPR at the edge. The FD pipeline detects and tracks the faces, extracts landmarks and quality of faces, to select appropriate faces for recognition and then sends them to face recognition server. LPR algorithm consecutively performs detection and recognition on the embedded platform. Extended YOLO model is utilized for face selection while pruned YOLO and LPRNet models are exploited for license plate detection and license plate reading, respectively. In order to enable real-time performance with high accuracy; optimized AI-models and software architecture are used. As a result of this study, we obtain a high-performance, high-precision and real-time combined face/LPR recognition system which can be very useful for surveillance and security applications.

show abstract

Section: Resultsmentioning

confidence: 99%

Simultaneous license plate recognition and face detection application at the edge

Alim

Kavakli

Okcu

et al. 2023

AI and Optical Data Sciences IV

View full text Add to dashboard Cite

show abstract

“…Additionally, the key modification in stem block structure and the block in the architecture were introduced to make it for the face and face landmark detection. It was not only achieved high accuracy than the state-of-the-art techniques, but also perform faster [37]. The YOLO5Face model can accurately detect the centre of the left and right eye, the tip of the nose, and the right and left mouth corners.…”

Section: Roi Selection and Trackingmentioning

confidence: 92%

“…The RGB image sequences have more detailed information compared to the thermal images and have a large database and pre-trained models for the detection of facial features. Haarcascade classifiers [35] for nose detection, multi-task cascaded convolution neural network (MTCNN) [36] and, more recently, the YOLO5Face detection [37] model are mostly used for the detection of faces and facial keypoints. The YOLO5Face detection model is designed by the YOLOv5 object detector [38].…”

Section: Roi Selection and Trackingmentioning

confidence: 99%

“…The TD-based method is simple to implement and performs efficiently in terms of object variation, scene variation, and the number of targets. In this work, the deep learning-based tracking by detection approach is used, which includes the YOLO5Face detector [37] and the dlib correlation tracker [40] to perform the automatic ROI selection in more challenging conditions, e.g., rapid movement, multiple people, and high dynamic background. Also, it handles the deviation of objects from the FOV or occlusions by reinitialization of the detectors.…”

Section: Roi Selection and Trackingmentioning

confidence: 99%

See 1 more Smart Citation

Non-contact respiratory rate monitoring using thermal and visible imaging: a pilot study on neonates

Maurya

Zwiggelaar

Chawla

et al. 2022

J Clin Monit Comput

View full text Add to dashboard Cite

Respiratory rate (RR) monitoring is essential in neonatal intensive care units. Despite its importance, RR is still monitored intermittently by manual counting instead of continuous monitoring due to the risk of skin damage with prolonged use of contact electrodes in preterm neonates and false signals due to displacement of electrodes. Thermal imaging has recently gained significance as a non-contact method for RR detection because of its many advantages. However, due to the lack of information in thermal images, the selection and tracking of the region of interest (ROI) in thermal images for neonates are challenging. This paper presents the integration of visible (RGB) and thermal (T) image sequences for the selection and tracking of ROI for breathing rate extraction. The deep-learning based tracking-by-detection approach is employed to detect the ROI in the RGB images, and it is mapped to the thermal images using the RGB-T image registration. The mapped ROI in thermal spectrum sequences gives the respiratory rate. The study was conducted first on healthy adults in different modes, including steady, motion, talking, and variable respiratory order. Subsequently, the method is tested on neonates in a clinical settings. The findings have been validated with a contact-based reference method.The average absolute error between the proposed and belt-based contact method in healthy adults reached 0.1 bpm and for more challenging conditions was approximately 1.5 bpm and 1.8 bpm, respectively. In the case of neonates, the average error is 1.5 bpm, which are promising results. The Bland–Altman analysis showed a good agreement of estimated RR with the reference method RR and this pilot study provided the evidence of using the proposed approach as a contactless method for the respiratory rate detection of neonates in clinical settings.

show abstract

“…Visual features. We detected the bounding boxes of faces on every frame with YOLO5Face [56] and applied triangular smooth to reduce glitches. Since occlusion is still challenging for facial expression recognition (FER) [57], we detected the head pose with FSA-Net [58] to validate the reliability and reduce the impact of misclassification.…”

Section: Processing Multimodal Featuresmentioning

confidence: 99%

Anchorage: Visual Analysis of Satisfaction in Customer Service Videos via Anchor Events

Kwai¹,

Wang²,

Wang³

et al. 2023

Preprint

View full text Add to dashboard Cite

Delivering customer services through video communications has brought new opportunities to analyze customer satisfaction for quality management. However, due to the lack of reliable self-reported responses, service providers are troubled by the inadequate estimation of customer services and the tedious investigation into multimodal video recordings. We introduce Anchorage, a visual analytics system to evaluate customer satisfaction by summarizing multimodal behavioral features in customer service videos and revealing abnormal operations in the service process. We leverage the semantically meaningful operations to introduce structured event understanding into videos which help service providers quickly navigate to events of their interest. Anchorage supports a comprehensive evaluation of customer satisfaction from the service and operation levels and efficient analysis of customer behavioral dynamics via multifaceted visualization views. We extensively evaluate Anchorage through a case study and a carefully-designed user study. The results demonstrate its effectiveness and usability in assessing customer satisfaction using customer service videos. We found that introducing event contexts in assessing customer satisfaction can enhance its performance without compromising annotation precision. Our approach can be adapted in situations where unlabelled and unstructured videos are collected along with sequential records.

show abstract

YOLO5Face: Why Reinventing a Face Detector

Cited by 93 publications

References 33 publications

Simultaneous license plate recognition and face detection application at the edge

Simultaneous license plate recognition and face detection application at the edge

Non-contact respiratory rate monitoring using thermal and visible imaging: a pilot study on neonates

Anchorage: Visual Analysis of Satisfaction in Customer Service Videos via Anchor Events

Contact Info

Product

Resources

About