Nội dung được dịch bởi AI, chỉ mang tính chất tham khảo
Một phương pháp nhận diện khuôn mặt bị che khuất dựa trên biến đổi động từ hình ảnh đến lớp sử dụng chỉ số tương đồng cấu trúc
Tóm tắt
Nhận diện khuôn mặt trong các môi trường không kiểm soát là một vấn đề thách thức trong thị giác máy tính do sự che khuất, độ nghiêng và sự thay đổi ánh sáng. Trong khi các kỹ thuật học máy có thể giải quyết việc nhận diện khuôn mặt bị che khuất, chúng yêu cầu phải huấn luyện lại khi cập nhật hình ảnh trong bộ sưu tập. Kỹ thuật Biến đổi Hình ảnh đến Lớp Động (DICW) cung cấp khả năng nhận diện theo thời gian thực mà không cần huấn luyện, duy trì thứ tự tự nhiên của các đặc điểm khuôn mặt (trán, mắt, mũi, miệng và cằm) để tránh những gián đoạn do sự che khuất gây ra. DICW tách biệt các mảnh khuôn mặt và tích hợp chúng thành một chuỗi có thứ tự thông qua quét raster. Nó tính toán khoảng cách từ hình ảnh tới lớp giữa các khuôn mặt truy vấn và mục tiêu bằng cách sử dụng các đường biến đổi tối ưu dọc theo các chiều thời gian và trong lớp. Bài báo này đề xuất một phương pháp cải tiến cho nhận diện khuôn mặt sử dụng DICW và chỉ số Tương đồng cấu trúc (SSIM), giảm thiểu sự biến đổi trong ánh sáng và độ tương phản để phù hợp thông tin cấu trúc. Một kỹ thuật nhận diện khuôn mặt tự động từ các chuỗi video với DICW cũng được trình bày. Các thực nghiệm trên Cơ sở Dữ liệu Khuôn mặt AR, Cơ sở Dữ liệu Chokepoint và các chuỗi video trong môi trường không kiểm soát cho thấy phương pháp đề xuất cải thiện đáng kể tỷ lệ nhận diện cho các hình ảnh bị che khuất. Phương pháp đề xuất đạt được sự cải thiện khoảng 5-6% trong tất cả các trường hợp đã xem xét so với các kỹ thuật tiên tiến khác.
Từ khóa
#nhận diện khuôn mặt #biến đổi động từ hình ảnh đến lớp #chỉ số tương đồng cấu trúc #che khuất #thị giác máy tínhTài liệu tham khảo
Golwalkar R, N Mehendale (2022) Masked-face recognition using deep metric learning and facemasknet-21. Appl Intell 1–12
R Szeliski (2022) Computer vision: algorithms and applications. Springer Nature
Jin M, Yu L, Zhou K, Yi Q (2022) Occlusion tolerant object recognition using visual memory selection model. Appl Intell 1–25
Li N, X Zhao (2022) A multi-modal dataset for gait recognition under occlusion. Appl Intell 1–18
P Viola, M Jones (2001) Rapid object detection using a boosted cascade of simple features. In: Proceedings of the 2001 IEEE computer society conference on computer vision and pattern recognition. CVPR 2001, volume 1. Ieee, p I–I
M Storer, M Urschler, H Bischof (2010) Occlusion detection for icao compliant facial photographs. In: 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition-Workshops. IEEE, 122–129
D Lin, X Tang (2007) Quality-driven face occlusion detection and recovery. In: 2007 IEEE Conference on Computer Vision and Pattern Recognition. IEEE, p 1–7
R Min, A Hadid, J-L Dugelay (2011) Improving the recognition of faces occluded by facial accessories. In: 2011 IEEE international conference on automatic face & gesture recognition (FG). IEEE, p 442–447
H Jia, AM Martinez (2008) Face recognition with occlusions in the training and testing sets. In: 2008 8th IEEE international conference on automatic face & gesture recognition. IEEE, p 1–6
X Wei, C-T Li, Y Hu (2012) Robust face recognition under varying illumination and occlusion considering structured sparsity. In: 2012 International Conference on Digital Image Computing Techniques and Applications (DICTA). IEEE, p 1–7
Liao S, Jain AK, Li SZ (2012) Partial face recognition: Alignment-free approach. IEEE Trans Pattern Anal Mach Intell 35(5):1193–1205
Tan X, Chen S, Zhou Z-H (2009) J Liu Face recognition under occlusions and variant expressions with partial similarity. IEEE Trans Inf Forensics Secur 4(2):217–230
Martinez AM (2002) Recognizing imprecisely localized partially occluded, and expression variant faces from a single sample per class. IEEE Trans Pattern Anal Mach Intell 24(6):748–763
Tan X, Chen S, Zhou Z-H (2005) F Zhang Recognizing partially occluded, expression variant faces from single training image per person with som and soft k-nn ensemble. IEEE Trans Neural Netw 16(4):875–886
Wei X, Li C-T, Hu Y (2013) Face recognition with occlusion using dynamic image-to-class warping (dicw)
X Wei, C-T Li, Y Hu (2013) Robust face recognition with occlusions in both reference and query images. In: 2013 International Workshop on Biometrics and Forensics(IWBF). IEEE p 1–4
Tzimiropoulos G, Zafeiriou S, Pantic M (2012) Subspace learning from image gradient orientations. IEEE Trans Pattern Anal Mach Intell 34(12):2454–2466
Zhu J, Cao D, Liu S, Lei Z, Li SZ (2012) Discriminant analysis with gabor phase for robust face recognition. In: 2012 5th IAPR International Conference on Biometrics (ICB). IEEE, p 13–18
A Morelli Andrés, S Padovani, M Tepper, J Jacobo-Berlles. Face recognition on partially occluded images using compressed sensing. Pattern Recognit Lett 36:235–242, 2014
Ramamonjisoa M, Du Y, Lepetit V (2020) Predicting sharp and accurate occlusion boundaries in monocular depth estimation using displacement fields. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. p 14648–14657
Huang S-M, Yang J-F (2012) Subface hidden markov models coupled with a universal occlusion model for partially occluded face recognition. IET Biometrics 1(3):149–159
G Suvarna Kumar, P Reddy, M Srinadh Swamy, S Gupta, et al. Skin based occlusion detection and face recognition using machine learning techniques. Int J Comput Appl 41(18), 2012
Yuan Y, Chu J, Leng L, Miao J, Kim B-G (2020) A scale-adaptive object-tracking algorithm with occlusion detection. EURASIP J Image Video Process 2020:1–15
He M, Zhang J, Shan S, Kan M, Chen X (2020) Deformable face net for pose invariant face recognition. Pattern Recognit 100:107113
Amutha R et al. (2012) A novel approach to face recognition under various facial expressions, occlusion and tilt angles. In: 2012 International Conference on Emerging Trends in Science, Engineering and Technology (INCOSET). IEEE, p 143–149
Venkat I, Khader AT, Subramanian KG, De Wilde P (2013) Recognizing occluded faces by exploiting psychophysically inspired similarity maps. Pattern Recognit Lett 34(8):903–911
Lin J, Ming J, Crookes D (2011) Robust face recognition with partial occlusion, illumination variation and limited training data by optimal feature selection. IET Comput Vis 5(1):23–32
A Vijayalakshmi, P Raj. An efficient method to recognize human faces from video sequences with occlusion. World Comput Sci Inform Technol J 5(2), 2015
Wei X, Li C-T, Hu Y (2013) Face recognition with occlusion using dynamic image-to-class warping (dicw). In: 2013 10th IEEE international conference and workshops on automatic face and gesture recognition (FG). IEEE, p 1–6
Ding R-X, Huang H, Shang J (2015) Patch-based locality-enhanced collaborative representation for face recognition. IET Image Process 9(3):211–217
Martinez AM (2002) Recognizing imprecisely localized, partially occluded, and expression variant faces from a single sample per class. IEEE Trans Pattern Anal Mach Intell 24(6):748–763
Jiang X, Lai J (2014) Sparse and dense hybrid representation via dictionary decomposition for face recognition. IEEE Trans Pattern Anal Mach Intell 37(5):1067–1079
Liao S, Jain AK, Li SZ (2012) Partial face recognition: Alignment-free approach. IEEE Trans Pattern Anal Mach Intell 35(5):1193–1205
Deng W, Hu J, Guo J (2012) Extended src: Undersampled face recognition via intraclass variant dictionary. IEEE Trans Pattern Anal Mach Intell 34(9):1864–1870
He R, Zheng W-S, Hu B-G (2010) Maximum correntropy criterion for robust face recognition. IEEE Trans Pattern Anal Mach Intell 33(8):1561–1576
N Zhang, M Paluri, Y Taigman, R Fergus, L Bourdev (2015) Beyond frontal faces: Improving person recognition using multiple cues. In: Proceedings of the IEEE conference on computer vision and pattern recognition. p 4804–4813
M Yang, L Zhang, J Yang, D Zhang (2011) Robust sparse coding for face recognition. In: CVPR 2011. IEEE, p 625–632
Wu CY, Ding JJ (2018) Occluded face recognition using low-rank regression with generalized gradient direction. Pattern Recognit 80:256–268
Cen F, Zhao X, Li W, Wang G (2021) Deep feature augmentation for occluded image classification. Pattern Recognit 111:107737
Zheng W, Gou C, Wang F-Y (2020) A novel approach inspired by optic nerve characteristics for few-shot occluded face recognition. Neurocomputing 376:25–41
Ou W, Luan X, Gou J, Zhou Q, Xiao W, Xiong X, Zeng W (2018) Robust discriminative nonnegative dictionary learning for occluded face recognition. Pattern Recognit Lett 107:41–49
Cen F, Wang G (2019) Dictionary representation of deep features for occlusion-robust face recognition. IEEE Access 7:26595–26605
H Qiu, D Gong, Z Li, W Liu, D Tao (2021) End2end occluded face recognition by masking corrupted features. IEEE Trans Pattern Anal Mach Intell
Li W, Chai Z, Tang Z (2021) A decomposition-based multi-objective immune algorithm for feature selection in learning to rank. Knowl-Based Syst 234:107577
Yu X, Tian X (2022) A fault detection algorithm for pipeline insulation layer based on immune neural network. Int J Press Vessels Pip 196:104611
Yu X, Ye X, Zhang S (2022) Floating pollutant image target extraction algorithm based on immune extremum region. Dig Signal Process 123:103442
R Spijkerman (2021) Immune inspired face recognition utilizing anomaly detection and symbiotic agents. University of Johannesburg (South Africa)
R Zhang, P Isola, Efros AA, Shechtman E, O Wang (2018) The unreasonable effectiveness of deep features as a perceptual metric. In: Proceedings of the IEEE conference on computer vision and pattern recognition. p 586–595
Li Q, He H, Lai H, Cai T, Wang Q, Gao QX (2022) Enhanced nuclear norm based matrix regression for occluded face recognition. Pattern Recognit 126:108585
He M, Zhang J, Shan S, Liu X, Wu Z, Chen X (2021) Locality-aware channel-wise dropout for occluded face recognition. IEEE Trans Image Process 31:788–798
Wei X, Li C-T, Lei Z, Yi D, Li SZ (2014) Dynamic image-to-class warping for occluded face recognition. IEEE Trans Inform Forens Sec 9(12):2035–2050
Azimipour M, Zawadzki RJ, Gorczynska I, Migacz J, Werner JS, Jonnal RS (2018) Intraframe motion correction for raster-scanned adaptive optics images using strip-based cross-correlation lag biases. PLoS ONE 13(10):e0206052
Wang Z, Bovik AC, Sheikh HR, Simoncelli EP (2004) Image quality assessment: from error visibility to structural similarity. IEEE Trans Image Process 13(4):600–612
Ahonen T, Hadid A, Pietikainen M (2006) Face description with local binary patterns: Application to face recognition. IEEE Trans Pattern Anal Mach Intellig 28(12):2037–2041
Y Wong, S Chen, S Mau, C Sanderson, BC Lovell (2011) Patch-based probabilistic image quality assessment for face selection and improved video-based face recognition. In: CVPR 2011 WORKSHOPS. IEEE, p 74–81
A Martinez, R Benavente (1998) The ar face database: Cvc technical report, 24
M Lal, K Kumar, RH Arain, A Maitlo, SA Ruk, H Shaikh. Study of face recognition techniques: A survey. Int J Adv Comput Sci Appl 9(6), 2018
AL Machidon, OM Machidon, PL Ogrutan (2019) Face recognition using eigenfaces, geometrical pca approximation and neural networks. In: 2019 42nd international conference on telecommunications and signal processing (TSP). IEEE, p 80–83
D Duisembayev (2022) Feasibility of face recognition algorithms on edge
Hassaballah M, Alshazly HA, Ali AA (2019) Ear recognition using local binary patterns: A comparative experimental study. Expert Syst Appl 118:182–200
Kiani K, Rezaeirad S, Rastgoo R (2021) Hmm-based face recognition using svd and half of the face image. Model Simul Electr Electron Eng 1(2):45–50
Hu L, Cui J (2019) Digital image recognition based on fractional-order-pca-svm coupling algorithm. Measurement 145:150–159
Cheng E-J, Chou K-P, Rajora S, Jin B-H, Tanveer M, Lin C-T, Young K-Y, Lin W-C, Prasad M (2019) Deep sparse representation classifier for facial recognition and detection system. Pattern Recognit Lett 125:71–77
Chen M-J, Bovik AC (2011) Fast structural similarity index algorithm. J Real Time Image Process 6(4):281–287
Zeng D, Veldhuis R, Spreeuwers L (2021) A survey of face recognition techniques under occlusion. IET Biometric 10(6):581–606