Big Brother is watching: AI algorithm can find you in CCTV footage without using face recognition
Indian researchers have built an algorithm that would allow surveillance footage to be searched and filtered for the purpose of “person retrieval.”
At the moment, surveillance footage must be searched manually to find a person based on common descriptors like clothing and gender. These identifying factors are known as soft biometrics.
The team used deep learning - which goes beyond machine learning (where patterns are set into algorithms and require supervision) by incorporating ‘self-learning’- to train a convolutional neural network (CNN) to recognize soft biometrics using computer vision.
Basically, you can tell this AI some details about the person you’re looking for and it’ll scour whatever video you give it.— Raymond Koh (@KohRaymond) October 23, 2018
Imagine a situation where a person is reported missing after two days. One might ask for footage from cam…https://t.co/KTBRUtQfzqhttps://t.co/kfvLmokSgB
According to the study, the algorithm correctly identified “28 persons out of 41 in a very challenging dataset with soft biometric attributes.” The researchers say that, with some minor tweaks, accuracy could be improved substantially.
The new technology could be crucial in finding missing persons and dangerous criminals, but also raises ethical issues about surveillance. With no legislation in place to limit its potential, the technology could be utilized for nefarious means such as illegal surveillance of activists and journalists, or wide scale monitoring of civilians by oppressive governments.
IBM has been working on similar technology for years. The company secretly used NYPD CCTV footage to develop technology to identify people based on soft biometrics, The Intercept revealed in September.
Like this story? Share it with a friend!