By effectively integrating facial features from different modalities, the interference of external environmental changes on facial recognition can be reduced, thereby improving the effectiveness of facial recognition. Therefore, a multi-modal facial recognition algorithm based on deep residual networks is proposed. This algorithm uses the BEMD method to decompose facial images at multiple scales and obtain multimodal feature representations. These features are input into a deep residual network, which quickly outputs identification results after feature extraction, reinforcement, and selection. Experimental results have shown that this method can effectively represent the multimodal features of facial images, and the residual components do not contain facial information, demonstrating strong representation capabilities; At the same time, it can effectively recognize faces with different expressions and angles, with short response time, achieving fast face recognition and significant application effects.
ChenAn-ming, LinQun-xiong, LiuWei-qiang.Multi feature fusion for masked face recognition based on contrastive learning[J]. Application Research of Computers,2024, 41(1): 277-281, 287.
[4]
AbateA F, CimminoL, Lorenzo-NavarroJ. An ablation study on part-based face analysis using a multi-input convolutional neural network and semantic segmentation[J]. Pattern Recognition Letters, 2023, 173: 45-49.
LiYun-hong, LiuXing-rui, XieRong-rong,et al.Low-resolution face recognition based on super-resolution reconstruction and common feature subspace[J]. Journal of Northwest University (Natural Science Edition), 2023, 53(2): 241-247.
WangQi, QianWei-zhong, LeiHang, et al.3D face recognition based on key feature enhancement mechanism[J]. Journal of University of Electronic Science and Technology of China, 2024, 53(2): 252-258.
WangHai-yong, PanHai-tao. Face recognition method based on hybrid adaptive loss function[J]. Journal of Frontiers of Computer Science & Technology,2024, 18(6): 1627-1636.
LiXi, YangMing, DuYa-ting.Occlusion face recognition based on partition selection and gabor wavelet[J]. Microelectronics & Computer,2023, 40(5): 39-46.
[13]
KhanZ, BoraguleA, D'AuriolB J J M. Improvised contrastive loss for improved face recognition in open-set nature[J]. Pattern Recognition Letters, 2024, 180: 120-126.
[14]
JalalA S, TripathiR. Face recognition under large age gap using age face generation[J]. International Journal of Biometrics, 2023, 15(2): 233-250.
ZhaoXiao, ChenYong, LiYi-yang. Low-resolution face recognition based on super-resolution reconstruction[J]. Fire Control & Command Control, 2024, 49(3): 151-155, 164.
[17]
SardarA, UmerS, Rout RanjeetK R,et al. A secure face recognition for iot-enabled healthcare system[J]. ACM Transactions on Sensor Networks, 2023, 19(3): 1.1-1.23.
LiXiao-xin, DingWei-jie, FangYi,et al.Occluded face recognition based on deep image prior and robust Markov random field[J]. Computer Science,2024,51(7):244-256.
[20]
VenkateswarlalP, NittaG R, PrasadA. Retracted article: ensemble of texture and shape descriptors using support vector machine classification for face recognition[J]. Journal of Ambient Intelligence and Humanized Computing, 2024,15(): 4589-4602.
ZhangZhong-wei, ChenHao.Research on face recognition model based on edge computing[J]. Journal of China Academy of Electronics and Information Technology, 2023, 18(4): 363-371.