Team Belgium Wins Top Prize for Image Privacy-Protecting App
• NTT Research announced that a team from NTT Global in Belgium won first place in its hackathon on attribute-based encryption (ABE).
• The winning demonstration was an application for ABE that could protect images containing private or sensitive information.
• A panel of judges from NTT DATA, NTT Research and the NTT Service Innovation Laboratory evaluated the competing teams’ demos.
NTT Research Hackathon
NTT Research, Inc., a division of NTT (TYO:9432), recently held a hackathon on attribute-based encryption (ABE) at the NTT Research offices in Sunnyvale, Calif. Five teams from around the world participated and submitted their demonstrations for evaluation by a panel of judges from NTT DATA, NTT Research and the NTT Service Innovation Laboratory.
Winning Demonstration
The winning demonstration was created by Senior Software Engineer Pascal Mathis and Data Scientist Jean-Philippe Cabay, both of NTT in Belgium. Titled “Confidentiality in Images,” this groundbreaking application for ABE allows for sharing data based on policies and attributes of users while protecting images containing private or sensitive information such as building logos, faces and license plates. Honorable mention went to demos created by teams from Italy and Romania.
Functionality
The demo involves three parts: 1) detecting and labeling objects; 2) encrypting images with mapped labels to ABE policies; 3) storing objects, metadata and blurred images in a database. This creates an extract, transfer load (ETL) pipeline which can ensure fine-grained control over who has access to specific information contained within images.
NTT R&D Forum Presentation Video
The Belgian team delivered a presentation video on their application during the virtual business conference – the NTT R&D Forum – which took place Nov 16-18th. The other competing teams also presented videos showcasing their work during this event.
Conclusion
Overall, this hackathon showcased impressive applications of ABE that can help balance privacy and security when it comes to shared data contained within images online.