A5022927890- Face recognition and analysis
- Generative Adversarial Networks and Image Synthesis
- Biometric Identification and Security
- Digital Media Forensic Detection
- Advanced Steganography and Watermarking Techniques
- Face and Expression Recognition
- User Authentication and Security Systems
- Cutaneous Melanoma Detection and Management
- Video Surveillance and Tracking Methods
- Law in Society and Culture
Shanghai Jiao Tong University
2020-2023
The popularization of intelligent devices including smartphones and surveillance cameras results in more serious privacy issues. De-identification is regarded as an effective tool for visual protection with the process concealing or replacing identity information. Most existing de-identification methods suffer from some limitations since they mainly focus on are usually non-reversible. In this paper, we propose a personalized invertible method based deep generative model, where main idea...
Unprecedented video collection and sharing have exacerbated privacy concerns led to increasing interest in privacy-preserving tools. A satisfactory de-identification tool should be able remove sensitive identity information from face videos while maintaining useful for other identity-agnostic tasks. Meanwhile, it is necessary allow the authority inspect real when abnormal events are detected. Existing methods only focus on study of de-identification, lack desired recovery ability granting...
Face de-identification involves concealing the true identity of a face while retaining other facial characteristics. Current target-generic methods typically disentangle features in latent space, using adversarial training to balance privacy and utility. However, this pattern often leads trade-off between utility, space remains difficult explain. To address these issues, we propose IDeudemon, which employs "divide conquer" strategy protect preserve utility step by maintaining good...
The development of modern social media allows millions private photos to be uploaded and shared, which provides a wide range image acquisition but extremely threatens personal privacy. Face de-identification is treated as an important privacy protection tool in multimedia data processing by modifying identity information. Although there exist many traditional methods widely used hide sensitive information, they all fail balance the trade-off between utility qualitative quantitative manners...
As more and personal photos are shared tagged in social media, security privacy protection becoming an unprecedentedly focus of attention. Avoiding risks such as unintended verification, becomes increasingly challenging. To enable people to enjoy uploading without having consider these concerns, it is crucial study techniques that allow individuals limit the identity information leaked visual data. In this paper, we propose a novel hybrid model consists two stages generate visually pleasing...
The widespread application of face recognition technology has exacerbated privacy threats. Face de-identification is an effective means protecting visual by concealing identity information. While deep learning-based methods have greatly improved results, most existing algorithms rely on 2D generative models that struggle to produce identity-consistent results for multiple views. In this paper, we focus disentanglement within the latest 3D-aware generation model, and propose advanced...
Advances in cameras and web technology have made it easy to capture share large amounts of face videos over an unknown audience with uncontrollable purposes. These raise increasing concerns about unwanted identity-relevant computer vision devices invading the characters's privacy. Previous de-identification methods rely on designing novel neural networks processing frame by frame, which ignore data feature redundancy continuity. Besides, these techniques are incapable well-balancing privacy...
Because of the explosive growth face photos as well their widespread dissemination and easy accessibility in social media, security privacy personal identity information becomes an unprecedented challenge. Meanwhile, convenience brought by advanced identity-agnostic computer vision technologies is attractive. Therefore, it important to use images while taking careful consideration protecting people's identities. Given a image, de-identification, also known anonymization, refers generating...