Using double attention for text tattoo localisation
Abstract Text tattoos contain rich information about an individual for forensic investigation. To extract this information, text tattoo localisation is the first and essential step. Previous tattoo studies applied existing object detectors to detect general tattoos, but none of them considered text...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-05-01
|
| Series: | IET Biometrics |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/bme2.12071 |
| _version_ | 1826820118186819584 |
|---|---|
| author | Xingpeng Xu Shitala Prasad Kuanhong Cheng Adams Wai Kin Kong |
| author_facet | Xingpeng Xu Shitala Prasad Kuanhong Cheng Adams Wai Kin Kong |
| author_sort | Xingpeng Xu |
| collection | DOAJ |
| description | Abstract Text tattoos contain rich information about an individual for forensic investigation. To extract this information, text tattoo localisation is the first and essential step. Previous tattoo studies applied existing object detectors to detect general tattoos, but none of them considered text tattoo localisation and they neglect the prior knowledge that text tattoos are usually inside or nearby larger tattoos and appear only on human skin. To use this prior knowledge, a prior knowledge‐based attention mechanism (PKAM) and a network named Text Tattoo Localisation Network based on Double Attention (TTLN‐DA) are proposed. In addition to TTLN‐DA, two variants of TTLN‐DA are designed to study the effectiveness of different prior knowledge. For this study, NTU Tattoo V2, the largest tattoo dataset and NTU Text Tattoo V1, the largest text tattoo dataset are established. To examine the importance of the prior knowledge and the effectiveness of the proposed attention mechanism and the networks, TTLN‐DA and its variants are compared with state‐of‐the‐art object detectors and text detectors. The experimental results indicate that the prior knowledge is vital for text tattoo localisation; The PKAM contributes significantly to the performance and TTLN‐DA outperforms the state‐of‐the‐art object detectors and scene text detectors. |
| first_indexed | 2024-03-09T08:10:36Z |
| format | Article |
| id | doaj.art-741bdcc7dda84a478416aadb9bf6f74b |
| institution | Directory Open Access Journal |
| issn | 2047-4938 2047-4946 |
| language | English |
| last_indexed | 2025-02-16T06:23:34Z |
| publishDate | 2022-05-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Biometrics |
| spelling | doaj.art-741bdcc7dda84a478416aadb9bf6f74b2025-02-03T06:47:38ZengWileyIET Biometrics2047-49382047-49462022-05-0111319921410.1049/bme2.12071Using double attention for text tattoo localisationXingpeng Xu0Shitala Prasad1Kuanhong Cheng2Adams Wai Kin Kong3School of Computer Science and Engineering Nanyang Technological University Singapore SingaporeVisual Intelligence Department/Advanced Manufacturing and Engineering Division Institute for Infocomm Research Agency for Science, Technology and Research Singapore SingaporeSchool of Physics and Optoelectronic Engineering Xidian University Xi’an ChinaSchool of Computer Science and Engineering Nanyang Technological University Singapore SingaporeAbstract Text tattoos contain rich information about an individual for forensic investigation. To extract this information, text tattoo localisation is the first and essential step. Previous tattoo studies applied existing object detectors to detect general tattoos, but none of them considered text tattoo localisation and they neglect the prior knowledge that text tattoos are usually inside or nearby larger tattoos and appear only on human skin. To use this prior knowledge, a prior knowledge‐based attention mechanism (PKAM) and a network named Text Tattoo Localisation Network based on Double Attention (TTLN‐DA) are proposed. In addition to TTLN‐DA, two variants of TTLN‐DA are designed to study the effectiveness of different prior knowledge. For this study, NTU Tattoo V2, the largest tattoo dataset and NTU Text Tattoo V1, the largest text tattoo dataset are established. To examine the importance of the prior knowledge and the effectiveness of the proposed attention mechanism and the networks, TTLN‐DA and its variants are compared with state‐of‐the‐art object detectors and text detectors. The experimental results indicate that the prior knowledge is vital for text tattoo localisation; The PKAM contributes significantly to the performance and TTLN‐DA outperforms the state‐of‐the‐art object detectors and scene text detectors.https://doi.org/10.1049/bme2.12071attention mechanismforensicstattoo localisationtext tattoo localisationvisual identification |
| spellingShingle | Xingpeng Xu Shitala Prasad Kuanhong Cheng Adams Wai Kin Kong Using double attention for text tattoo localisation IET Biometrics attention mechanism forensics tattoo localisation text tattoo localisation visual identification |
| title | Using double attention for text tattoo localisation |
| title_full | Using double attention for text tattoo localisation |
| title_fullStr | Using double attention for text tattoo localisation |
| title_full_unstemmed | Using double attention for text tattoo localisation |
| title_short | Using double attention for text tattoo localisation |
| title_sort | using double attention for text tattoo localisation |
| topic | attention mechanism forensics tattoo localisation text tattoo localisation visual identification |
| url | https://doi.org/10.1049/bme2.12071 |
| work_keys_str_mv | AT xingpengxu usingdoubleattentionfortexttattoolocalisation AT shitalaprasad usingdoubleattentionfortexttattoolocalisation AT kuanhongcheng usingdoubleattentionfortexttattoolocalisation AT adamswaikinkong usingdoubleattentionfortexttattoolocalisation |