Abstract
Real-time face detection has many challenges, such as non-frontal faces, tiny faces, occlusions, and multifarious backgrounds. Real-time face detection can be done by using Convolutional Neural Network (CNN) models, which result in elevated performance but have a huge computation time. It needs to be implemented on high-end computational devices to produce more accurate face detection results for high resolution images. We proposed a light architecture based on CNN for deep learning-based feature extraction and detection of human faces. The challenges faced during real-time face detection, such as occlusions, different scales, different backgrounds, varying positions, lighting, and poses, are resolved, and faces are detected accurately using the proposed framework. The amount of computation required for real-time face detection is reduced. This light architecture consists of two modules: the backbone module is used to contract the input size of the image and extract the features; the detection module transforms the feature map between prediction layers and detects faces at various scales. In our architecture, we use mini-inception blocks that minimize the computation cost and are implemented using available low-end system configurations without the need for external hardware. The proposed model uses anchor boxes to predict bounding boxes using dimensional clusters. The model is trained and tested using images from the WIDER Face dataset, which has images of various challenging conditions. Finally, images with multiple faces detected are displayed as output. The proposed work shows an increased accuracy rate with reduced computation cost over state-of-the-art performance on the benchmark dataset.
Similar content being viewed by others
References
Li Y, Zeng J, Shan S and Chen X 2019 Occlusion aware facial expression recognition using CNN with attention mechanism. IEEE Trans. Image Process. 28(5): 2439–2450
Hariri W and Farah N 2021 Recognition of 3D emotional facial expression based on handcrafted and deep feature combination. Pattern Recognit. Lett. 148: 84–91
Zheng G and Xu Y 2021 Efficient face detection and tracking in video sequences based on deep learning. Inf. Sci. 568: 265–285
Ullah N, Javed A, Ali Ghazanfar M, Alsufyani A and Bourouis S 2022 A novel DeepMaskNet model for face mask detection and masked facial recognition. J. King Saud Univ. Comput. Inf. Sci. 34(10): 9905–9914
Soni N, Sharma E K and Kapoor A 2021 Hybrid meta-heuristic algorithm based deep neural network for face recognition. J. Comput. Sci. 51: 101352
Viswanatha Reddy G, Dharma Savarni C V R and Mukherjee S 2020 Facial expression recognition in the wild, by fusion of deep learnt and hand-crafted features. Cogn. Syst. Res. 62: 23–34
Khan M Z, Harous S, Hassan S U, Ghani Khan M U, Iqbal R and Mumtaz S 2019 Deep unified model for face recognition based on convolution neural network and edge computing. IEEE Access 7: 72622–72633
Wu H, Zhang K and Tian G 2018 Simultaneous face detection and pose estimation using convolutional neural network cascade. IEEE Access 6: 49563–49575
Putro M D, Kurnianggoro L and Jo K-H 2021 High performance and efficient real-time face detector on central processing unit based on convolutional neural network. IEEE Trans. Ind. Inf. 17(7): 4449–4457
Zhang B, Li J, Wang Y, Tai Y, Wang C, Li J, Huang F, Xia Y, Pei W and Ji R 2020 ASFD: automatic and scalable face detector https://doi.org/10.48550/arXiv.2003.11228
Kumar K K, Kasiviswanadham Y, Indira D V S N V, Priyanka palesetti P and Bhargavi Ch V 2021 Criminal face identification system using deep learning algorithm multi-task cascade neural network (MTCNN). Mater. Today Proc.https://doi.org/10.1016/j.matpr.2021.06.373
Qi R, Jia R-S, Mao Q-C, Sun H-M and Zuo L-Q 2019 Face detection method based on cascaded convolutional networks. IEEE Access 7: 110740–110748
Li H, Li C, Li G and Chen L 2021 A real-time table grape detection method based on improved YOLOv4-tiny network in complex background. Biosyst. Eng. 212: 347–359
Gao J and Yang T 2022 Face detection algorithm based on improved TinyYOLOv3 and attention mechanism. Comput. Commun. 181: 329–337
Yang C-Y and Chen H H 2021 Efficient face detection in the fisheye image domain. IEEE Trans. Image Process. 30: 5641–5651
Shuang K, Lyu Z, Loo J and Zhang W 2021 Scale-balanced loss for object detection. Pattern Recognit. 117: 107997
Cai Y, Luan T, Gao H, Wang H, Chen L, Li Y, Sotelo M A and Li Z 2021 YOLOv4-5D: an effective and efficient object detector for autonomous driving. IEEE Trans. Instrum. Meas. 70: 1–13
Ryan C, O’Sullivan B, Elrasad A, Cahill A, Lemley J, Kielty P, Posch C and Perot E 2021 Real-time face & eye tracking and blink detection using event cameras. Neural Netw. 141: 87–97
Jin Q, Mu C, Tian L and Ran F 2020 A region generation based model for occluded face detection. Procedia Comput. Sci. 174: 454–462
Chou K-Y and Chen Y-P 2020 Real-time and low-memory multi-faces detection system design with Naive Bayes classifier implemented on FPGA. IEEE Trans. Circuits Syst. Video Technol. 30(11): 4380–4389
Li X, Yang Z and Wu H 2020 Face detection based on receptive field enhanced multi-task cascaded convolutional neural networks. IEEE Access 8: 174922–174930
Bochkovskiy A, Wang C-Y and Liao H-YM 2020 YOLOv4: Optimal Speed and Accuracy of Object Detection. https://doi.org/10.48550/arXiv.2004.10934
Luo J, Liu J, Lin J and Wang Z 2020 A lightweight face detector by integrating the convolutional neural network with the image pyramid. Pattern Recognit. Lett. 133: 180–187
Earp S W F, Noinongyao P, Cairns J A and Ganguly A 2019 Face detection with feature pyramids and landmarks. https://doi.org/10.48550/arXiv.1912.00596
Wu W, Yin Y, Wang X and Xu D 2019 Face detection with different scales based on faster R-CNN. IEEE Trans. Cybern. 49(11): 4017–4028
Zhang J, Wu X, Zhu J and Hoi S C H 2017 Feature agglomeration networks for single stage face detection. https://doi.org/10.48550/arXiv.1712.00721
Zeng D, Liu H, Zhao F, Ge S, Shen W and Zhang Z 2019 Proposal pyramid networks for fast face detection. Inf. Sci. 495: 136–149
Zhang S, Wang X, Lei Z and Li S Z 2019 Faceboxes: a CPU real-time and accurate unconstrained face detector. Neurocomputing 364: 297–309
Guo G, Wang H, Yan Y, Zheng J and Li B 2020 A fast face detection method via convolutional neural network. Neurocomputing 395: 128–137
Yu B and Tao D 2019 Anchor cascade for efficient face detection. IEEE Trans. Image Process. 28(5): 2490–2501
Zhang Z, Shen W, Qiao S, Wang Y, Wang B and Yuille A 2018 Robust face detection via learning small faces on hard images. https://doi.org/10.48550/arXiv.1811.11662
Sun X, Wu P and Hoi S C H 2018 Face detection using deep learning: an improved faster RCNN approach. Neurocomputing 299: 42–50
Attia A and Dayan S 2018 Detecting and counting tiny faces. https://doi.org/10.48550/arXiv.1801.06504
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Lakshmanan, B., Vaishnavi, A., Ananthapriya, R. et al. A novel deep facenet framework for real-time face detection based on deep learning model. Sādhanā 48, 265 (2023). https://doi.org/10.1007/s12046-023-02329-3
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s12046-023-02329-3