Unsupervised detection of contrast enhanced highlight landmarks
Abstract In the field of landmark detection based on deep learning, most of the research utilise convolutional neural networks to represent landmarks, and rarely adopt Transformer to represent and encode landmarks. Meanwhile, many works focus on modifying the network structure to improve network per...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2023-10-01
|
| Series: | IET Computer Vision |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/cvi2.12197 |
| _version_ | 1797663021031489536 |
|---|---|
| author | Tao Wu Wenzhuo Fan Shuxian Li Qingqing Li Jianlin Zhang Meihui Li |
| author_facet | Tao Wu Wenzhuo Fan Shuxian Li Qingqing Li Jianlin Zhang Meihui Li |
| author_sort | Tao Wu |
| collection | DOAJ |
| description | Abstract In the field of landmark detection based on deep learning, most of the research utilise convolutional neural networks to represent landmarks, and rarely adopt Transformer to represent and encode landmarks. Meanwhile, many works focus on modifying the network structure to improve network performance, and there is little research on the distribution of landmarks. In this article,the authors propose an unsupervised model to extract landmarks of objects in images. First, Transformer structure is combined with the convolutional neural network structure to represent and encode the landmarks; next, positive and negative sample pairs between landmarks are constructed, so that the semantically consistent landmarks on the image are pulled closer in the feature space and the semantically inconsistent landmarks are pushed farther in the feature space; then the authors concentrate their attention on the most active points to distinguish the landmarks of an object from the background; finally, based on the new contrastive loss, the network reconstructs the image by the landmarks of the object that are continuously learnt during training. Experiments show that the proposed model achieves better performance than other unsupervised methods on the CelebA, Annotated Facial Landmarks in the Wild, 300W datasets. |
| first_indexed | 2024-03-11T19:08:32Z |
| format | Article |
| id | doaj.art-bf87f964cea94dc08ae3ac211e91296e |
| institution | Directory Open Access Journal |
| issn | 1751-9632 1751-9640 |
| language | English |
| last_indexed | 2024-03-11T19:08:32Z |
| publishDate | 2023-10-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Computer Vision |
| spelling | doaj.art-bf87f964cea94dc08ae3ac211e91296e2023-10-10T04:15:41ZengWileyIET Computer Vision1751-96321751-96402023-10-0117780481310.1049/cvi2.12197Unsupervised detection of contrast enhanced highlight landmarksTao Wu0Wenzhuo Fan1Shuxian Li2Qingqing Li3Jianlin Zhang4Meihui Li5Key Laboratory of Optical Engineering Chinese Academy of Sciences Chengdu ChinaSchool of Electronic, Electrical and Communication Engineering University of Chinese Academy of Sciences Beijing ChinaKey Laboratory of Optical Engineering Chinese Academy of Sciences Chengdu ChinaInstitute of Optics and Electronics Chinese Academy of Sciences Chengdu ChinaInstitute of Optics and Electronics Chinese Academy of Sciences Chengdu ChinaInstitute of Optics and Electronics Chinese Academy of Sciences Chengdu ChinaAbstract In the field of landmark detection based on deep learning, most of the research utilise convolutional neural networks to represent landmarks, and rarely adopt Transformer to represent and encode landmarks. Meanwhile, many works focus on modifying the network structure to improve network performance, and there is little research on the distribution of landmarks. In this article,the authors propose an unsupervised model to extract landmarks of objects in images. First, Transformer structure is combined with the convolutional neural network structure to represent and encode the landmarks; next, positive and negative sample pairs between landmarks are constructed, so that the semantically consistent landmarks on the image are pulled closer in the feature space and the semantically inconsistent landmarks are pushed farther in the feature space; then the authors concentrate their attention on the most active points to distinguish the landmarks of an object from the background; finally, based on the new contrastive loss, the network reconstructs the image by the landmarks of the object that are continuously learnt during training. Experiments show that the proposed model achieves better performance than other unsupervised methods on the CelebA, Annotated Facial Landmarks in the Wild, 300W datasets.https://doi.org/10.1049/cvi2.12197computer visionimage processingunsupervised learning |
| spellingShingle | Tao Wu Wenzhuo Fan Shuxian Li Qingqing Li Jianlin Zhang Meihui Li Unsupervised detection of contrast enhanced highlight landmarks IET Computer Vision computer vision image processing unsupervised learning |
| title | Unsupervised detection of contrast enhanced highlight landmarks |
| title_full | Unsupervised detection of contrast enhanced highlight landmarks |
| title_fullStr | Unsupervised detection of contrast enhanced highlight landmarks |
| title_full_unstemmed | Unsupervised detection of contrast enhanced highlight landmarks |
| title_short | Unsupervised detection of contrast enhanced highlight landmarks |
| title_sort | unsupervised detection of contrast enhanced highlight landmarks |
| topic | computer vision image processing unsupervised learning |
| url | https://doi.org/10.1049/cvi2.12197 |
| work_keys_str_mv | AT taowu unsuperviseddetectionofcontrastenhancedhighlightlandmarks AT wenzhuofan unsuperviseddetectionofcontrastenhancedhighlightlandmarks AT shuxianli unsuperviseddetectionofcontrastenhancedhighlightlandmarks AT qingqingli unsuperviseddetectionofcontrastenhancedhighlightlandmarks AT jianlinzhang unsuperviseddetectionofcontrastenhancedhighlightlandmarks AT meihuili unsuperviseddetectionofcontrastenhancedhighlightlandmarks |