One-shot learning for facial sketch recognition using the siamese convolutional neural network
Deep Convolutional Neural Networks have been widely used in computer vision tasks like classifying an image and detecting an object within an image. To archive the state-of-the-art performance, it normally requires a huge number of labeled samples. However, in the facial sketch recognition tasks, co...
| id |
Ahmad Sabri N.I.; Setumin S. |
|---|---|
| spelling |
Ahmad Sabri N.I.; Setumin S. 2-s2.0-85107708964 One-shot learning for facial sketch recognition using the siamese convolutional neural network 2021 ISCAIE 2021 - IEEE 11th Symposium on Computer Applications and Industrial Electronics 10.1109/ISCAIE51753.2021.9431773 https://www.scopus.com/inward/record.uri?eid=2-s2.0-85107708964&doi=10.1109%2fISCAIE51753.2021.9431773&partnerID=40&md5=425183158589ce2fda3837175ae6e0f2 Deep Convolutional Neural Networks have been widely used in computer vision tasks like classifying an image and detecting an object within an image. To archive the state-of-the-art performance, it normally requires a huge number of labeled samples. However, in the facial sketch recognition tasks, collecting this amount of samples is not feasible. Each subject will only have one sketch and one photo. To address this, a One-shot Learning method with Siamese Network is proposed in this paper due to the fact that it only requires one training sample per class. The network comprises two identical model instances that share the same architecture and weights to be trained to learn the similarity between the two images. The similarity score is computed by using Euclidean distance. Some four different activation functions are evaluated in this research to see how feasible those functions to be used in this recognition task. The results demonstrate that the most suitable activation function for this task is sigmoid, with an accuracy of 100% after about 300 learning iterations for 10- way One-shot Learning. The evaluation is extended to the CUHK dataset and the results indicate the same accuracy pattern. © 2021 IEEE. Institute of Electrical and Electronics Engineers Inc. English Conference paper |
| author |
2-s2.0-85107708964 |
| spellingShingle |
2-s2.0-85107708964 One-shot learning for facial sketch recognition using the siamese convolutional neural network |
| author_facet |
2-s2.0-85107708964 |
| author_sort |
2-s2.0-85107708964 |
| title |
One-shot learning for facial sketch recognition using the siamese convolutional neural network |
| title_short |
One-shot learning for facial sketch recognition using the siamese convolutional neural network |
| title_full |
One-shot learning for facial sketch recognition using the siamese convolutional neural network |
| title_fullStr |
One-shot learning for facial sketch recognition using the siamese convolutional neural network |
| title_full_unstemmed |
One-shot learning for facial sketch recognition using the siamese convolutional neural network |
| title_sort |
One-shot learning for facial sketch recognition using the siamese convolutional neural network |
| publishDate |
2021 |
| container_title |
ISCAIE 2021 - IEEE 11th Symposium on Computer Applications and Industrial Electronics |
| container_volume |
|
| container_issue |
|
| doi_str_mv |
10.1109/ISCAIE51753.2021.9431773 |
| url |
https://www.scopus.com/inward/record.uri?eid=2-s2.0-85107708964&doi=10.1109%2fISCAIE51753.2021.9431773&partnerID=40&md5=425183158589ce2fda3837175ae6e0f2 |
| description |
Deep Convolutional Neural Networks have been widely used in computer vision tasks like classifying an image and detecting an object within an image. To archive the state-of-the-art performance, it normally requires a huge number of labeled samples. However, in the facial sketch recognition tasks, collecting this amount of samples is not feasible. Each subject will only have one sketch and one photo. To address this, a One-shot Learning method with Siamese Network is proposed in this paper due to the fact that it only requires one training sample per class. The network comprises two identical model instances that share the same architecture and weights to be trained to learn the similarity between the two images. The similarity score is computed by using Euclidean distance. Some four different activation functions are evaluated in this research to see how feasible those functions to be used in this recognition task. The results demonstrate that the most suitable activation function for this task is sigmoid, with an accuracy of 100% after about 300 learning iterations for 10- way One-shot Learning. The evaluation is extended to the CUHK dataset and the results indicate the same accuracy pattern. © 2021 IEEE. |
| publisher |
Institute of Electrical and Electronics Engineers Inc. |
| issn |
|
| language |
English |
| format |
Conference paper |
| accesstype |
|
| record_format |
scopus |
| collection |
Scopus |
| _version_ |
1828987871211552768 |