Copyright © All rights reserved micansinfotech.com
Photo sharing is an attractive feature which popularizes Online Social Networks (OSNs). Unfortunately, it may leak users’ privacy if they are allowed to post, comment, and tag a photo freely.we attempt to address this issue and study the scenario when a user shares a photo containing individuals other than himself/herself (termed co-photo for short). To prevent possible privacy leakage of a photo, we design a mechanism to enable each individual in a photo be aware of the posting activity and participate in the decision making on the photo posting. For this purpose, we need an efficient facial recognition (FR) system that can recognize everyone in the photo. However, more demanding privacy setting may limit the number of the photos publicly available to train the FR system. To deal with this dilemma, our mechanism attempts to utilize users’ private photos to design a personalized FR system specifically trained to differentiate possible photo co-owners without leaking their privacy. We also develop a distributed consensus based method to reduce the computational complexity and protect the private training set. We show that our system is superior to other possible approaches in terms of recognition ratio and efficiency. Our mechanism is implemented as a proof of concept Android application on Facebook’s platform. The power-law distribution is caused by the preferential attach process, in which the probability of a user A connecting to a user B is proportional to the number of B’s existing connections. show the snapshots of the contact network and fan network in YA, respectively. We see some nodes do not have either fans or contacts, while a few nodes have a very large degree.
Users care about privacy are unlikely to put photos online. Perhaps it is exactly those people who really want to have a photo privacy protection scheme. To break this dilemma, we propose a privacy-preserving distributed collaborative training system as our FR engine. In our system, we ask each of our users to establish a private photo set of their own.
We use these private photos to build personal FR engines based on the specific social context and promise that during FR training, only the discriminating rules are revealed but nothing else With the training data (private photo sets) distributed among users, this problem could be formulated as a typical secure multi-party computation problem. Intuitively, we may apply cryptographic technique to protect the private photos, but the computational and communication cost may pose a serious problem for a large OSN.
They define a pair wise conditional random field (CRF) model to find the optimal joint labeling by maximizing the conditional density. Specifically, they use the existing labeled photos as the training samples and combine the photo co occurrence statistics and baseline FR score to improve the accuracy of face annotation. discuss the difference between the traditional FR system and the FR system that is designed specifically for OSNs.
They point out that a customized FR system for each user is expected to be much more accurate in his/her own photo collections. social networks such as Face book. Unfortunately, careless photo posting may reveal privacy of individuals in a posted photo. To curb the privacy leakage, we proposed to enable individuals potentially in a photo to give the permissions before posting a co-photo. We designed a privacy-preserving FR system to identify individuals in a co-photo.
According to our protocol, her friends only communicate with her and they have no idea of what they are computing for. Friend list could also be revealed during the classifier reuse stage. For example, suppose Alice want to find ubt between Bob and Tom, which has already been computed by Bob.
Alice will first query user k to see if ukj has already been computed. If this query is made in plaintext, Bob immediately knows Alice and Bob are friends. To address this problem, Alice will first make a list for desired classifiers use private set operations to query against her neighbors’ classifiers lists one by one.
To break this dilemma, we propose a privacy-preserving distributed collaborative training system as our FR engine. In our system, we ask each of our users to establish a private photo set of their own. We use these private photos to build personal FR engines based on the specific social context and promise that during FR training; only the discriminating rules are revealed but nothing else.
Propose to use multiple personal FR engines to work collaboratively to improve the recognition ratio. Specifically, they use the social context to select the suitable FR engines that contain the identity of the queried face image with high probability This data isolation property is the essence of our secure collaborative learning model and the detailed security analysis.
A survey was conducted in to study the effectiveness of the existing countermeasure of un tagging and shows that this countermeasure is far from satisfactory users are worrying about offending their friends when un tagging. As a result, they provide a tool to enable users to restrict others from seeing their photos when posted as a complementary strategy to protect privacy.
However, this method will introduce a large number of manual tasks for end users. In, propose a game-theoretic scheme in which the privacy policies are collaboratively enforced over the shared data. This happens when the appearance of user I have changed or the photos in the training set are modified adding new images or deleting existing images. The friendship graph may change over time.
In propose a game-theoretic scheme in which the privacy policies are collaboratively enforced over the shared data. Basically, in our proposed one-against-one strategy a user needs to establish classifiers between self, friend and friend, friend also known as the two loops in Algorithm.
During the first loop, there are no privacy concerns of Alice’s friend list because friendship graph is undirected. However, in the second loop, Alice need to coordinate all her friends to build classifiers between them. According to our protocol, her friends only communicate with her and they have no idea of what they are computing.
Photo sharing is one of the most popular features in online social networks such as Facebook. Unfortunately, careless photo posting may reveal privacy of individuals in a posted photo. To curb the privacy leakage, we proposed to enable individuals potentially in a photo to give the permissions before posting a co-photo. We designed a privacy-preserving FR system to identify individuals in a co-photo.
The proposed system is featured with low computation cost and confidentiality of the training set. Theoretical analysis and experiments were conducted to show effectiveness and efficiency of the proposed scheme. We expect that our proposed scheme be very useful in protecting users’ privacy in photo/image sharing over online social networks.