More Like this

1. Sign Language Video Anonymization

   Xia, Zhaoyang; Chen, Yuxiao; Zhangli, Qilong; Huenerfauth, Matt; Neidle, Carol; Metaxas, Dimitris (June 2022, LREC2022 10th Workshop on the Representation and Processing of Sign Languages: Multilingual Sign Language Resources)

   Deaf signers who wish to communicate in their native language frequently share videos on the Web. However, videos cannot preserve privacy—as is often desirable for discussion of sensitive topics—since both hands and face convey critical linguistic information and therefore cannot be obscured without degrading communication. Deaf signers have expressed interest in video anonymization that would preserve linguistic content. However, attempts to develop such technology have thus far shown limited success. We are developing a new method for such anonymization, with input from ASL signers. We modify a motion-based image animation model to generate high-resolution videos with the signer identity changed, but with preservation of linguistically significant motions and facial expressions. An asymmetric encoder-decoder structured image generator is used to generate the high-resolution target frame from the low-resolution source frame based on the optical flow and confidence map. We explicitly guide the model to attain clear generation of hands and face by using bounding boxes to improve the loss computation. FID and KID scores are used for evaluation of the realism of the generated frames. This technology shows great potential for practical applications to benefit deaf signers. 

   more » « less
2. Sign Language Video Anonymization

   Xia, Zhaoyang; Chen, Yuxiao; Zhangli, Qilong; Huenerfauth, Matt; Neidle, Carol; Metaxas, Dimitri (January 2022, Proceedings of the LREC2022 10th Workshop on the Representation and Processing of Sign Languages: Multilingual Sign Language Resources)
3. Sign Language Video Anonymization

   Xia, Zhaoyang; Chen, Yuxiao; Zhangli, Qilong; Huenerfauth, Matt; Neidle, Carol; Metaxas, Dimitris (June 2022, Proceedings of the LREC2022 10th Workshop on the Representation and Processing of Sign Languages: Multilingual Sign Language Resources, Marseille, France, 25 June 2022)

   Deaf signers who wish to communicate in their native language frequently share videos on the Web. However, videos cannot preserve privacy—as is often desirable for discussion of sensitive topics—since both hands and face convey critical linguistic information and therefore cannot be obscured without degrading communication. Deaf signers have expressed interest in video anonymization that would preserve linguistic content. However, attempts to develop such technology have thus far shown limited success. We are developing a new method for such anonymization, with input from ASL signers. We modify a motion-based image animation model to generate high-resolution videos with the signer identity changed, but with preservation of linguistically significant motions and facial expressions. An asymmetric encoder-decoder structured image generator is used to generate the high-resolution target frame from the low-resolution source frame based on the optical flow and confidence map. We explicitly guide the model to attain clear generation of hands and face by using bounding boxes to improve the loss computation. FID and KID scores are used for evaluation of the realism of the generated frames. This technology shows great potential for practical applications to benefit deaf signers. 

   more » « less
4. Fraud De-Anonymization for Fun and Profit

   https://doi.org/10.1145/3243734.3243770  

   Hernandez, Nestor; Rahman, Mizanur; Recabarren, Ruben; Carbunar, Bogdan (October 2018, Proceedings of the 2018 ACM SIGSAC Conference on Computer and Communications Security)

   The persistence of search rank fraud in online, peer-opinion systems, made possible by crowdsourcing sites and specialized fraud workers, shows that the current approach of detecting and filtering fraud is inefficient. We introduce a fraud de-anonymization approach to disincentivize search rank fraud: attribute user accounts flagged by fraud detection algorithms in online peer-opinion systems, to the human workers in crowdsourcing sites, who control them. We model fraud de-anonymization as a maximum likelihood estimation problem, and introduce UODA, an unconstrained optimization solution. We develop a graph based deep learning approach to predict ownership of account pairs by the same fraudster and use it to build discriminative fraud de-anonymization (DDA) and pseudonymous fraudster discovery algorithms (PFD). To address the lack of ground truth fraud data and its pernicious impacts on online systems that employ fraud detection, we propose the first cheating-resistant fraud de-anonymization validation protocol, that transforms human fraud workers into ground truth, performance evaluation oracles. In a user study with 16 human fraud workers, UODA achieved a precision of 91%. On ground truth data that we collected starting from other 23 fraud workers, our co-ownership predictor significantly outperformed a state-of-the-art competitor, and enabled DDA and PFD to discover tens of new fraud workers, and attribute thousands of suspicious user accounts to existing and newly discovered fraudsters. 

   more » « less
5. AOT: Anonymization by oblivious transfer

   Javani, Farid; Sherman, Alan T. (January 2021, Privacy Enhanced Tecologies, submtted)

   null (Ed.)

   We introduce AOT, an anonymous communication system based on mix network architecture that uses oblivious transfer (OT) to deliver messages. Using OT to deliver messages helps AOT resist blending (n-1) attacks and helps AOT preserve receiver anonymity, even if a covert adversary controls all nodes in AOT. AOT comprises three levels of nodes, where nodes at each level perform a different function and can scale horizontally. The sender encrypts their payload and a tag---derived from a secret shared between the sender and receiver---with the public key of a Level-2 node and sends them to a Level-1 node. On a public bulletin board, Level-3 nodes publish tags associated with messages ready to be retrieved. Each receiver checks the bulletin board, identifies tags, and receives the associated messages using OT. A receiver can receive their messages even if the receiver is offline when messages are ready. Through what we call a ``handshake'' process, communicants can use the AOT protocol to establish shared secrets anonymously. Users play an active role in contributing to the unlinkability of messages: periodically, users initiate requests to AOT to receive dummy messages, such that an adversary cannot distinguish real and dummy requests. 

   more » « less