Few-Shot Personalized Saliency Prediction Based on Adaptive Image Selection Considering Object and Visual Attention
A few-shot personalized saliency prediction based on adaptive image selection considering object and visual attention is presented in this paper. Since general methods predicting personalized saliency maps (PSMs) need a large number of training images, the establishment of a theory using a small num...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-04-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/8/2170 |
| _version_ | 1827718878463000576 |
|---|---|
| author | Yuya Moroto Keisuke Maeda Takahiro Ogawa Miki Haseyama |
| author_facet | Yuya Moroto Keisuke Maeda Takahiro Ogawa Miki Haseyama |
| author_sort | Yuya Moroto |
| collection | DOAJ |
| description | A few-shot personalized saliency prediction based on adaptive image selection considering object and visual attention is presented in this paper. Since general methods predicting personalized saliency maps (PSMs) need a large number of training images, the establishment of a theory using a small number of training images is needed. To tackle this problem, although finding persons who have visual attention similar to that of a target person is effective, all persons have to commonly gaze at many images. Thus, it becomes difficult and unrealistic when considering their burden. On the other hand, this paper introduces a novel adaptive image selection (AIS) scheme that focuses on the relationship between human visual attention and objects in images. AIS focuses on both a diversity of objects in images and a variance of PSMs for the objects. Specifically, AIS selects images so that selected images have various kinds of objects to maintain their diversity. Moreover, AIS guarantees the high variance of PSMs for persons since it represents the regions that many persons commonly gaze at or do not gaze at. The proposed method enables selecting similar users from a small number of images by selecting images that have high diversities and variances. This is the technical contribution of this paper. Experimental results show the effectiveness of our personalized saliency prediction including the new image selection scheme. |
| first_indexed | 2024-03-10T20:31:02Z |
| format | Article |
| id | doaj.art-788122f74ea24a92950ff7835a2c66f0 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T20:31:02Z |
| publishDate | 2020-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-788122f74ea24a92950ff7835a2c66f02023-11-19T21:22:30ZengMDPI AGSensors1424-82202020-04-01208217010.3390/s20082170Few-Shot Personalized Saliency Prediction Based on Adaptive Image Selection Considering Object and Visual AttentionYuya Moroto0Keisuke Maeda1Takahiro Ogawa2Miki Haseyama3Graduate School of Information Science and Technology, Hokkaido University, N-14, W-9, Kita-ku, Sapporo, Hokkaido 060-0814, JapanOffice of Institutional Research, Hokkaido University, N-8, W-5, Kita-ku, Sapporo, Hokkaido 060-0808, JapanFaculty of Information Science and Technology, Hokkaido University, N-14, W-9, Kita-ku, Sapporo, Hokkaido 060-0814, JapanFaculty of Information Science and Technology, Hokkaido University, N-14, W-9, Kita-ku, Sapporo, Hokkaido 060-0814, JapanA few-shot personalized saliency prediction based on adaptive image selection considering object and visual attention is presented in this paper. Since general methods predicting personalized saliency maps (PSMs) need a large number of training images, the establishment of a theory using a small number of training images is needed. To tackle this problem, although finding persons who have visual attention similar to that of a target person is effective, all persons have to commonly gaze at many images. Thus, it becomes difficult and unrealistic when considering their burden. On the other hand, this paper introduces a novel adaptive image selection (AIS) scheme that focuses on the relationship between human visual attention and objects in images. AIS focuses on both a diversity of objects in images and a variance of PSMs for the objects. Specifically, AIS selects images so that selected images have various kinds of objects to maintain their diversity. Moreover, AIS guarantees the high variance of PSMs for persons since it represents the regions that many persons commonly gaze at or do not gaze at. The proposed method enables selecting similar users from a small number of images by selecting images that have high diversities and variances. This is the technical contribution of this paper. Experimental results show the effectiveness of our personalized saliency prediction including the new image selection scheme.https://www.mdpi.com/1424-8220/20/8/2170personalized saliency mapadaptive image selectionmulti-task CNNobject detection |
| spellingShingle | Yuya Moroto Keisuke Maeda Takahiro Ogawa Miki Haseyama Few-Shot Personalized Saliency Prediction Based on Adaptive Image Selection Considering Object and Visual Attention Sensors personalized saliency map adaptive image selection multi-task CNN object detection |
| title | Few-Shot Personalized Saliency Prediction Based on Adaptive Image Selection Considering Object and Visual Attention |
| title_full | Few-Shot Personalized Saliency Prediction Based on Adaptive Image Selection Considering Object and Visual Attention |
| title_fullStr | Few-Shot Personalized Saliency Prediction Based on Adaptive Image Selection Considering Object and Visual Attention |
| title_full_unstemmed | Few-Shot Personalized Saliency Prediction Based on Adaptive Image Selection Considering Object and Visual Attention |
| title_short | Few-Shot Personalized Saliency Prediction Based on Adaptive Image Selection Considering Object and Visual Attention |
| title_sort | few shot personalized saliency prediction based on adaptive image selection considering object and visual attention |
| topic | personalized saliency map adaptive image selection multi-task CNN object detection |
| url | https://www.mdpi.com/1424-8220/20/8/2170 |
| work_keys_str_mv | AT yuyamoroto fewshotpersonalizedsaliencypredictionbasedonadaptiveimageselectionconsideringobjectandvisualattention AT keisukemaeda fewshotpersonalizedsaliencypredictionbasedonadaptiveimageselectionconsideringobjectandvisualattention AT takahiroogawa fewshotpersonalizedsaliencypredictionbasedonadaptiveimageselectionconsideringobjectandvisualattention AT mikihaseyama fewshotpersonalizedsaliencypredictionbasedonadaptiveimageselectionconsideringobjectandvisualattention |