iPrivacy: Image Privacy Protection by Identifying Sensitive Objects via Deep Multi-Task Learning

By Jun Ye Yu, Baopeng Zhang, Zheng Kuang, Dan Lin, Jianping Fan. IEEE Transactions on Information Forensics and Security, 2017

322 citations recorded by the source corpus.

To achieve automatic recommendation of privacy settings for image sharing, a new tool called iPrivacy (image privacy) is developed for releasing the burden from users on setting the privacy preferences when they share their images for special moments. Specifically, this paper consists of the following contributions: 1) massive social images and their privacy settings are leveraged to learn the object-privacy relatedness effectively and identify a set of privacy-sensitive object classes automatically; 2) a deep multi-task learning algorithm is developed to jointly learn more representative deep convolutional neural networks and more discriminative tree classifier, so that we can achieve fast and accurate detection of large numbers of privacy-sensitive object classes; 3) automatic recommendation of privacy settings for image sharing can be achieved by detecting the underlying privacy-sensitive objects from the images being shared, recognizing their classes, and identifying their privacy settings according to the object-privacy relatedness; and 4) one simple solution for image privacy protection is provided by blurring the privacy-sensitive objects automatically. We have conducted extensive experimental studies on real-world images and the results have demonstrated both the efficiency and effectiveness of our proposed approach.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Sociology and Political Science

View Paper: https://www.semanticscholar.org/paper/39f90ed6cdb6bfcbd7b80035b0c760c2301cbe72

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Qibin Hou
2. Deng-Ping Fan
3. Ming-Ming Cheng
4. Jiaxing Zhao
5. Yun Liu
6. Yang Xiang
7. Ming Ding
8. Farhad Farokhi

In Computer Vision and Pattern Recognition

1. Deng-Ping Fan
2. Ming-Ming Cheng
3. Qibin Hou
4. Ling Shao
5. Yun Liu
6. Jiaxing Zhao
7. Jianbing Shen
8. Huchuan Lu

In Sociology and Political Science

1. Ming Ding
2. Farhad Farokhi
3. Bo Liu
4. Zihuai Lin
5. Wanlei Zhou
6. Wenny Rahayu
7. Tianqing Zhu
8. Sina Shaham

Citations in the corpus

• Multi-view low-rank sparse subspace clustering (2017, background): “Opposed to [35, 40, 46], the proposed approach can deal with highly heterogeneous multi-view data coming from different modalities.”
• A Survey of Recent Advances in CNN-based Single Image Crowd Counting and Density Estimation (2017, unknown): “Inspired by the superior results achieved by simultaneous learning of related tasks [76, 104], Sindagi et al. [92] and Mars-den et al. [66] explored multi-task learning to boost individual task performance.”
• Rethinking RGB-D Salient Object Detection: Models, Data Sets, and Large-Scale Benchmarks (2019, background): “, stereo estimating/matching [110], multiple salient person detection, salient instance detection [19], sensitive object detection [111], image segmentation [112]).”
• Adversaries or allies? Privacy and deep learning in big data era (2018, methodology): “For example, Yu et al 7 developed an approach to automate the process of identifying privacy sensitive object classes and their privacy settings.”
• When Machine Learning Meets Privacy (2020, methodology): “raining datasets can be the target of privacy attacks, leading to sensitive information leakage. Meanwhile, researchers have also tried to use ML for privacy protection. As an example, the authors of [174] have developed a method for automatic recognition of privacy-sensitive object classes and adju”