- Baidu said that the face-scanning model will use artificial intelligence
- The model was trained on a dataset of 100,000 faces
According to a report by Technode, Baidu has released an open-source tool to detect if individuals in crowds are wearing face masks. This comes when authorities in China have taken big steps to curb the spread of Covid-19.
The report says that many cities like Guangzhou and Beijing are enforcing the use of masks in public areas like restaurants, shopping malls, and public transport. Till now, mask inspections have been mainly manual, making it difficult to identify non-mask wearers in crowds
According to the report, Baidu said that the face-scanning model will use artificial intelligence to identify people in real-time who are not wearing masks or those who are wearing them incorrectly. It added that the system can identify non-mask wearers with 96.5 per cent accuracy. Developers will need a small amount of data to train the tool for their own use.
Baidu also added that the model was trained on a dataset of 100,000 faces. The company claims that the system is first of its kind and can help businesses check if their employees are wearing masks.