Lightweight facial landmark detection network based on improved MobileViT

被引:1
|
作者
Song, Limei [1 ]
Hong, Chuanfei [1 ]
Gao, Tian [1 ]
Yu, Jiali [1 ]
机构
[1] Tiangong Univ, Tianjin Key Lab Intelligent Control Elect Equipmen, Tianjin, Peoples R China
关键词
Facial landmark; MobileViT; Lightweight; Transformer;
D O I
10.1007/s11760-023-02975-4
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
The long-distance link between facial landmarks cannot be modeled by the current CNN-based facial landmark detection networks, and these networks typically have many parameters that consume substantial computational resources. This paper proposes a multi-scale lightweight facial landmark detection network with CNN and Transformer multi-branch parallelism. Based on MobileViT, the network incorporates MobileOne Block and simplified Ghost BottleNeck lightweight network structure. Compared to MobileViT on the WFLW dataset, the number of network parameters is reduced by 49.18%, the failure rate is reduced by 3.20%, the detection speed is improved by 41.73%, the FLOPS is reduced by 64.83%, and the NME is improved by 0.45% and 1.31% on the test and pose subsets, respectively. The data proves that the global information extraction of facial landmarks is more accurate after adding the Transformer structure. This paper also compares with other networks, and the result shows that improved MobileViT achieves more accurate detection with fewer model parameters.
引用
收藏
页码:3123 / 3131
页数:9
相关论文
共 50 条
  • [1] Lightweight facial landmark detection network based on improved MobileViT
    Limei Song
    Chuanfei Hong
    Tian Gao
    Jiali Yu
    [J]. Signal, Image and Video Processing, 2024, 18 : 3123 - 3131
  • [2] Facial Landmark Configuration for Improved Detection
    Huang, C.
    Efraty, B. A.
    Kurkure, U.
    Papadakis, M.
    Shah, S. K.
    Kakadiaris, I. A.
    [J]. 2012 IEEE INTERNATIONAL WORKSHOP ON INFORMATION FORENSICS AND SECURITY (WIFS), 2012, : 13 - 18
  • [3] LIGHTWEIGHT FACIAL LANDMARK DETECTION WITH WEAKLY SUPERVISED LEARNING
    Lai, Shenqi
    Liu, Lei
    Chai, Zhenhua
    Wei, Xiaolin
    [J]. 2021 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA & EXPO WORKSHOPS (ICMEW), 2021,
  • [4] Lightweight Rail Fastener Fault Detection Method Based on Improved MobileVit and Grad-Cam
    Yao D.
    Wang Y.
    Yang J.
    Xu L.
    Zhang F.
    [J]. Zhendong Ceshi Yu Zhenduan/Journal of Vibration, Measurement and Diagnosis, 2024, 44 (01): : 52 - 59and196
  • [5] ELFLN: An Efficient Lightweight Facial Landmark Network Based on Hybrid Knowledge Distillation
    Chen, Shidong
    Wang, Yalun
    Bian, Huicong
    Lu, Qin
    [J]. PATTERN RECOGNITION AND COMPUTER VISION, PRCV 2023, PT VIII, 2024, 14432 : 485 - 497
  • [6] Improved Stacked Hourglass Network with Offset Learning for Robust Facial Landmark Detection
    Shi, Husen
    Wang, Zengfu
    [J]. 2019 9TH INTERNATIONAL CONFERENCE ON INFORMATION SCIENCE AND TECHNOLOGY (ICIST2019), 2019, : 58 - 64
  • [7] High-Precision and Lightweight Facial Landmark Detection Algorithm
    Xu Lihuai
    Li Zhe
    Jiang Jiajia
    Duan Fajie
    Fu Xiao
    [J]. LASER & OPTOELECTRONICS PROGRESS, 2020, 57 (21)
  • [8] High-Precision and Lightweight Facial Landmark Detection Algorithm
    Xu Lihuai
    Li Zhe
    Jiang Jiajia
    Duan Fajie
    Fu Xiao
    [J]. LASER & OPTOELECTRONICS PROGRESS, 2020, 57 (24)
  • [9] Recurrent neural network for facial landmark detection
    Chen, Yu
    Yang, Jian
    Qian, Jianjun
    [J]. NEUROCOMPUTING, 2017, 219 : 26 - 38
  • [10] Style Aggregated Network for Facial Landmark Detection
    Dong, Xuanyi
    Yan, Yan
    Ouyang, Wanli
    Yang, Yi
    [J]. 2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, : 379 - 388