Self-Improved Learning for Salient Object Detection
-
Published:2023-12-04
Issue:23
Volume:13
Page:12966
-
ISSN:2076-3417
-
Container-title:Applied Sciences
-
language:en
-
Short-container-title:Applied Sciences
Author:
Li Songyuan1ORCID, Zeng Hao1, Wang Huanyu1, Li Xi1
Affiliation:
1. College of Computer Science and Technology, Zhejiang University, Hangzhou 310027, China
Abstract
Salient Object Detection (SOD) aims at identifying the most visually distinctive objects in a scene. However, learning a mapping directly from a raw image to its corresponding saliency map is still challenging. First, the binary annotations of SOD impede the model from learning the mapping smoothly. Second, the annotator’s preference introduces noisy labeling in the SOD datasets. Motivated by these, we propose a novel learning framework which consists of the Self-Improvement Training (SIT) strategy and the Augmentation-based Consistent Learning (ACL) scheme. SIT aims at reducing the learning difficulty, which provides smooth labels and improves the SOD model in a momentum-updating manner. Meanwhile, ACL focuses on improving the robustness of models by regularizing the consistency between raw images and their corresponding augmented images. Extensive experiments on five challenging benchmark datasets demonstrate that the proposed framework can play a plug-and-play role in various existing state-of-the-art SOD methods and improve their performances on multiple benchmarks without any architecture modification.
Subject
Fluid Flow and Transfer Processes,Computer Science Applications,Process Chemistry and Technology,General Engineering,Instrumentation,General Materials Science
Reference65 articles.
1. Mahadevan, V., and Vasconcelos, N. (2009, January 20–25). Saliency-based discriminant tracking. Proceedings of the IEEE Conference of Computer Vision and Pattern Recognition (CVPR), Miami Beach, FL, USA. 2. Fang, H., Gupta, S., Iandola, F.N., Srivastava, R.K., Deng, L., Dollár, P., Gao, J., He, X., Mitchell, M., and Platt, J.C. (2015, January 7–12). From captions to visual concepts and back. Proceedings of the IEEE Conference of Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA. 3. Study of Saliency in Objective Video Quality Assessment;Zhang;IEEE Trans. Image Process.,2017 4. Zhao, R., Ouyang, W., and Wang, X. (2013, January 25–27). Unsupervised Salience Learning for Person Re-identification. Proceedings of the IEEE Conference of Computer Vision and Pattern Recognition (CVPR), Portland, OR, USA. 5. Liu, G., and Fan, D. (2013, January 7–8). A Model of Visual Attention for Natural Image Retrieval Computing Companion. Proceedings of the International Conference on Information Science and Cloud, Guangzhou, China.
|
|