期刊
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
卷 -, 期 -, 页码 10066-10076出版社
IEEE COMPUTER SOC
DOI: 10.1109/CVPR46437.2021.00994
关键词
-
资金
- National Natural Science Foundation of China [61871325]
- National Key Research and Development Program of China [2018AAA0102803]
- CSIRO's Machine Learning and Artificial Intelligence Future Science Platform (MLAI FSP)
- Swiss National Science Foundation via the Sinergia [CRSII5-180359]
- Swiss National Science Foundation (SNF) [CRSII5_180359] Funding Source: Swiss National Science Foundation (SNF)
This paper proposes a method to enhance the detection ability of visual salient object detection and camouflaged object detection by leveraging contradictory information. By using easy positive samples in the COD dataset to improve SOD model robustness, introducing a similarity measure module, and employing an adversarial learning network to handle labeling uncertainty, the proposed solution achieves state-of-the-art performance for both tasks.
Visual salient object detection (SOD) aims at finding the salient object(s) that attract human attention, while camouflaged object detection (COD) on the contrary intends to discover the camouflaged object(s) that hidden in the surrounding. In this paper, we propose a paradigm of leveraging the contradictory information to enhance the detection ability of both salient object detection and camouflaged object detection. We start by exploiting the easy positive samples in the COD dataset to serve as hard positive samples in the SOD task to improve the robustness of the SOD model. Then, we introduce a similarity measure module to explicitly model the contradicting attributes of these two tasks. Furthermore, considering the uncertainty of labeling in both tasks' datasets, we propose an adversarial learning network to achieve both higher order similarity measure and network confidence estimation. Experimental results on benchmark datasets demonstrate that our solution leads to state-of-the-art (SOTA) performance for both tasks(1).
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据