Abstract:Images today are increasingly shared online on social networking sites such as Facebook, Flickr, Foursquare, and Instagram. Despite that current social networking sites allow users to change their privacy preferences, this is often a cumbersome task for the vast majority of users on the Web, who face difficulties in assigning and managing privacy settings. Thus, automatically predicting images' privacy to warn users about private or sensitive content before uploading these images on social networking sites has become a necessity in our current interconnected world. In this paper, we explore learning models to automatically predict appropriate images' privacy as private or public using carefully identified image-specific features. We study deep visual semantic features that are derived from various layers of Convolutional Neural Networks (CNNs) as well as textual features such as user tags and deep tags generated from deep CNNs. Particularly, we extract deep (visual and tag) features from four pre-trained CNN architectures for object recognition, i.e., AlexNet, GoogLeNet, VGG-16, and ResNet, and compare their performance for image privacy prediction. Results of our experiments on a Flickr dataset of over thirty thousand images show that the learning models trained on features extracted from ResNet outperform the state-of-the-art models for image privacy prediction. We further investigate the combination of user tags and deep tags derived from CNN architectures using two settings: (1) SVM on the bag-of-tags features; and (2) text-based CNN. Our results show that even though the models trained on the visual features perform better than those trained on the tag features, the combination of deep visual features with image tags shows improvements in performance over the individual feature sets.

Leveraging Content Sensitiveness and User Trustworthiness to Recommend Fine-Grained Privacy Settings for Social Image Sharing

Cross-Network Social User Embedding with Hybrid Differential Privacy Guarantees

Deep Learning-Based User Privacy Settings Recommendation in Online Social Networks

Recessive Social Networking: Preventing Privacy Leakage against Reverse Image Search

SMPCS: Self-Managed Privacy Control and Sharing Scheme Based on Thumbnail-Preserving and ECDH for Social Networks

A Personal Privacy Preserving Framework

Privacy Intelligence: A Survey on Image Privacy in Online Social Networks

Towards Privacy-Preserving and Practical Image-Centric Social Discovery

Enabling Privacy-Preserving Image-Centric Social Discovery

Comparative Evaluations of Privacy on Digital Images

Privacy-aware Location Privacy Preference Recommendations

IEye: Personalized Image Privacy Detection

A machine learning based approach for user privacy preservation in social networks

Towards Automated Content-based Photo Privacy Control in User-Centered Social Networks.

Enhancing User-Centric Privacy Protection: An Interactive Framework through Diffusion Models and Machine Unlearning

Image Privacy Prediction Using Deep Neural Networks

SocialGuard: An adversarial example based privacy-preserving technique for social images

Secure Outsourcing Image Feature Extraction

Privacy-preserving outsourcing of image global feature detection

Protecting Privacy in Shared Photos Via Adversarial Examples Based Stealth

Trust-Aware Media Recommendation in Heterogeneous Social Networks