Chinese researchers scrape masked photos to train facial recognition systems
A dataset of photographs of people wearing ma sks developed by Chinese researchers drew criticism for its data collection practices and potential uses. R esearchers at Wuhan University collected photographs of 5,000 masked faces of 525 people , ostensibly to help train facial recognition systems that would identify individuals wearing face masks as a means of controlling the COVID-19 pandemic . The researchers said the images are of 'public figures' gathered 'from massive internet resources , ' but refused to discuss how they chose the people added to the dataset or whether or not this may impact their privacy. Critics argued that the photographs were collected without the consent of their owners or people appearing in them, raising privacy concerns. The dataset also prompted questions about potential uses of the dataset, not least in a Chinese context - including the monitoring of Uyghurs in Xinjiang province. System 🤖 Real-World Masked Face dataset O perator: Developer: Wuhan University Country: China Sector: Health; Research/academia Purpose: Improve facial recognition algorithms Technology: Dataset; Facial recognition; Computer vision Issue: Privacy; Ethics/values; Dual/multi-use; Surveillance
- Date it happened
- 2020-01-01
- Product, system or model
- Real World Masked Face Dataset
This incident was imported from AIAAIC and is used under CC BY-SA 4.0. Our additions to it — the structured fields, the translation, the checks against other reports — are published under the same licence.
This is a record of what was reported, not a finding that anyone broke the law. If it names your organisation and you believe it is wrong, the corrections process is free and open to everyone.