Historical Blurry Video-Based Face Recognition
Face recognition is a widely used computer vision, which plays an increasingly important role in user authentication systems, security systems, and consumer electronics. The models for most current applications are based on high-definition digital cameras. In this paper, we focus on digital images d...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-09-01
|
| Series: | Journal of Imaging |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2313-433X/10/9/236 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841560681535504384 |
|---|---|
| author | Lujun Zhai Suxia Cui Yonghui Wang Song Wang Jun Zhou Greg Wilsbacher |
| author_facet | Lujun Zhai Suxia Cui Yonghui Wang Song Wang Jun Zhou Greg Wilsbacher |
| author_sort | Lujun Zhai |
| collection | DOAJ |
| description | Face recognition is a widely used computer vision, which plays an increasingly important role in user authentication systems, security systems, and consumer electronics. The models for most current applications are based on high-definition digital cameras. In this paper, we focus on digital images derived from historical motion picture films. Historical motion picture films often have poorer resolution than modern digital imagery, making face detection a more challenging task. To approach this problem, we first propose a trunk–branch concatenated multi-task cascaded convolutional neural network (TB-MTCNN), which efficiently extracts facial features from blurry historical films by combining the trunk with branch networks and employing various sizes of kernels to enrich the multi-scale receptive field. Next, we build a deep neural network-integrated object-tracking algorithm to compensate for failed recognition over one or more video frames. The framework combines simple online and real-time tracking with deep data association (Deep SORT), and TB-MTCNN with the residual neural network (ResNet) model. Finally, a state-of-the-art image restoration method is employed to reduce the effect of noise and blurriness. The experimental results show that our proposed joint face recognition and tracking network can significantly reduce missed recognition in historical motion picture film frames. |
| format | Article |
| id | doaj-art-11cc91697590454a853459a39fb4f308 |
| institution | Kabale University |
| issn | 2313-433X |
| language | English |
| publishDate | 2024-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Imaging |
| spelling | doaj-art-11cc91697590454a853459a39fb4f3082025-01-03T15:52:49ZengMDPI AGJournal of Imaging2313-433X2024-09-0110923610.3390/jimaging10090236Historical Blurry Video-Based Face RecognitionLujun Zhai0Suxia Cui1Yonghui Wang2Song Wang3Jun Zhou4Greg Wilsbacher5Electrical and Computer Engineering Department, Prairie View A&M University, Prairie View, TX 77446, USAElectrical and Computer Engineering Department, Prairie View A&M University, Prairie View, TX 77446, USAComputer Science Department, Prairie View A&M University, Prairie View, TX 77446, USAComputer Science and Engineering Department, University of South Carolina, Columbia, SC 29425, USAComputer Science and Engineering Department, University of South Carolina, Columbia, SC 29425, USAMoving Image Research Collections, University Libraries, University of South Carolina, Columbia, SC 29425, USAFace recognition is a widely used computer vision, which plays an increasingly important role in user authentication systems, security systems, and consumer electronics. The models for most current applications are based on high-definition digital cameras. In this paper, we focus on digital images derived from historical motion picture films. Historical motion picture films often have poorer resolution than modern digital imagery, making face detection a more challenging task. To approach this problem, we first propose a trunk–branch concatenated multi-task cascaded convolutional neural network (TB-MTCNN), which efficiently extracts facial features from blurry historical films by combining the trunk with branch networks and employing various sizes of kernels to enrich the multi-scale receptive field. Next, we build a deep neural network-integrated object-tracking algorithm to compensate for failed recognition over one or more video frames. The framework combines simple online and real-time tracking with deep data association (Deep SORT), and TB-MTCNN with the residual neural network (ResNet) model. Finally, a state-of-the-art image restoration method is employed to reduce the effect of noise and blurriness. The experimental results show that our proposed joint face recognition and tracking network can significantly reduce missed recognition in historical motion picture film frames.https://www.mdpi.com/2313-433X/10/9/236face detectionface recognitionface trackinghistorical blurry video |
| spellingShingle | Lujun Zhai Suxia Cui Yonghui Wang Song Wang Jun Zhou Greg Wilsbacher Historical Blurry Video-Based Face Recognition Journal of Imaging face detection face recognition face tracking historical blurry video |
| title | Historical Blurry Video-Based Face Recognition |
| title_full | Historical Blurry Video-Based Face Recognition |
| title_fullStr | Historical Blurry Video-Based Face Recognition |
| title_full_unstemmed | Historical Blurry Video-Based Face Recognition |
| title_short | Historical Blurry Video-Based Face Recognition |
| title_sort | historical blurry video based face recognition |
| topic | face detection face recognition face tracking historical blurry video |
| url | https://www.mdpi.com/2313-433X/10/9/236 |
| work_keys_str_mv | AT lujunzhai historicalblurryvideobasedfacerecognition AT suxiacui historicalblurryvideobasedfacerecognition AT yonghuiwang historicalblurryvideobasedfacerecognition AT songwang historicalblurryvideobasedfacerecognition AT junzhou historicalblurryvideobasedfacerecognition AT gregwilsbacher historicalblurryvideobasedfacerecognition |