Toward Automated Online Photo Privacy

Squicciarini, Anna; Caragea, Cornelia; Balakavi, Rahul

doi:10.1145/2983644

Cited by 34 publications

(35 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…GIST [Oliva and Torralba 2001] encodes global descriptors for images and extracts a set of perceptual dimensions (naturalness, openness, roughness, expansion, and ruggedness) that represent the dominant spatial structure of the scene. Squicciarini et al [2014Squicciarini et al [ , 2017a performed an in-depth analysis of image privacy classification using Flickr images and found that SIFT and image tags work best for predicting privacy of users' images.…”

Section: Related Workmentioning

confidence: 99%

Image Privacy Prediction Using Deep Neural Networks

Tonge

Caragea

2020

ACM Trans. Web

Self Cite

View full text Add to dashboard Cite

Images today are increasingly shared online on social networking sites such as Facebook, Flickr, Foursquare, and Instagram. Image sharing occurs not only within a group of friends but also more and more outside a user's social circles for purposes of social discovery. 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. When these privacy settings are used inappropriately, online image sharing can potentially lead to unwanted disclosures and privacy violations. 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. Among all four networks, we observe that ResNet produces the best feature representations for this task. We also fine-tune the pre-trained CNN architectures on our privacy dataset and compare their performance with the models trained on pre-trained features. The results show that even though the overall performance obtained using the fine-tuned networks is comparable to that of pre-trained networks, the fine-tuned networks provide an improved performance for the private class as compared to models trained on the pre-trained features. 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. We compare these models with the models trained on ResNet visual features obtained for privacy prediction. Our results show that even though the models trained

show abstract

Section: Related Workmentioning

confidence: 99%

Image Privacy Prediction Using Deep Neural Networks

Tonge

Caragea

2020

ACM Trans. Web

Self Cite

View full text Add to dashboard Cite

show abstract

“…Privacy in online social networks have been studied in depth. A group of approaches predict the privacy labels of content that are about to be shared online [10]. Those approaches assume that the privacy of content does not change based on the context.…”

Section: Related Workmentioning

confidence: 99%

TURP: Managing Trust for Regulating Privacy in Internet of Things

Kökciyan

Yolum

2020

IEEE Internet Comput.

View full text Add to dashboard Cite

Internet of Things [IoT] applications, such as smart home or ambient assisted living systems, promise useful services to end users. Most of these services rely heavily on sharing and aggregating information among devices; many times raising privacy concerns. Contrary to traditional systems, where privacy of each user is managed through well-defined policies, the scale, dynamism, and heterogeneity of the IoT systems make it impossible to specify privacy policies for all possible situations. Alternatively, this paper argues that handling of privacy has to be reasoned by the IoT devices, depending on the norms, context, as well as the trust among entities. We present a technique, where an IoT device collects information from others, evaluates the trustworthiness of the information sources to decide the suitability of sharing information with others. We demonstrate the applicability of the technique over an IoT pilot study.

show abstract

“…Psychological acceptability of protection mechanisms has been largely recognized as a main challenge in the security research community since the seminal work of Saltzer and Schroeder [1975]. For instance, policy specification has been recognized as a difficult and costly task for end-users [Fang and LeFevre 2010;Klemperer et al 2012;Squicciarini et al 2017]. This is even more challenging in community-centered systems mainly due to the multi-party nature of these systems, wherein the members of the community should collaboratively specify protection requirements for shared resources.…”

Section: Community-centered Collaborative Systemsmentioning

confidence: 99%

“…For instance, Yu et al [2017] use deep multi-task learning to automatically detect privacy-sensitive objects in images and recommend privacy settings for their protection. A number of approaches (e.g., [Zerr et al 2012;Squicciarini et al 2015;Squicciarini et al 2017]) leverage both visual features of images and other information for privacy-aware image classification and the generation of privacy settings. For instance, Squicciarini et al [2015] propose an adaptive policy prediction system based on association rule mining to generate personalized privacy settings based on social context and personal traits as well as on image content and metadata.…”

Section: Usability and Transparencymentioning

confidence: 99%

Survey on Access Control for Community-Centered Collaborative Systems

2018

Self Cite

View full text Add to dashboard Cite

The last decades have seen a growing interest and demand for community-centered collaborative systems and platforms. These systems and platforms aim to provide an environment in which users can collaboratively create, share and manage resources. While offering attractive opportunities for online collaboration and information sharing, they also open several security and privacy issues. This has attracted several research efforts towards the design and implementation of novel access control solutions that can handle the complexity introduced by collaboration. Despite these efforts, transition to practice has been hindered by the lack of maturity of the proposed solutions. The access control mechanisms typically adopted by commercial collaborative systems like online social network websites and collaborative editing platforms, are still rather rudimentary and do not provide users with a sufficient control over their resources. This survey examines the growing literature on access control for collaborative systems centered on communities, and identifies the main challenges to be addressed in order to facilitate the adoption of collaborative access control solutions in real-life settings. Based on the literature study, we delineate a roadmap for future research in the area of access control for community-centered collaborative systems.

show abstract

Toward Automated Online Photo Privacy

Cited by 34 publications

References 44 publications

Image Privacy Prediction Using Deep Neural Networks

Image Privacy Prediction Using Deep Neural Networks

TURP: Managing Trust for Regulating Privacy in Internet of Things

Survey on Access Control for Community-Centered Collaborative Systems

Contact Info

Product

Resources

About