Face Recognition
Face recognition is a computer vision technology that performs identity verification and identification based on facial biometric features. It captures facial images or video, applies algorithms to extract facial feature points, and compares them against information stored in a database to confirm identity or retrieve individuals. The technology is widely applied in security surveillance, access control, financial payment, and device unlocking, and is characterized by contactless operation and rapid response.
Core Facts
Timeline (last 90 days)
Face recognition accuracy is usually above 95% in good lighting, and may decline in low light or at extreme angles.
Machine learning can be used to detect spam, recognize faces, and determine plant diseases from leaf images.
In a facial recognition system with 99.9% accuracy, if there are only 10 real targets among 1 million people, approximately 99% of the roughly 1,010 alerts generated would be false positives.
There can be a vast gap between a technology's performance in a laboratory setting and its actual effectiveness in real, complex, target-sparse public environments.
Facial data is a biometric identifier that is nearly impossible to change — passwords can be reset, but faces cannot be replaced.
Airports and Madison Square Garden in New York are using facial recognition technology.
The greatest advantage of voiceprint recognition over facial recognition is that it requires no additional hardware and can be collected by any device with a microphone.
In the field of face recognition, a low-discriminability feature space corresponds to a distribution where both intra-class variance and inter-class variance are small, causing Euclidean distance or cosine similarity metrics to fail.
Facial recognition systems have significantly higher misidentification rates for people with darker skin than for those with lighter skin.
Airports have in recent years introduced large numbers of technologies such as self-service check-in, facial recognition, and automated security screening.
2 more timeline events
All Facts (13)
Research by Joy Buolamwini at the MIT Media Lab found that several commercial facial recognition systems had a misidentification rate as high as 34.7% for dark-skinned women, compared to only 0.8% for light-skinned men
75%Unverified当前许多平台的人脸识别系统仍然依赖2D视频验证，容易被AI换脸技术突破
80%UnverifiedFace recognition accuracy is usually above 95% in good lighting, and may decline in low light or at extreme angles.
50%UnverifiedMachine learning can be used to detect spam, recognize faces, and determine plant diseases from leaf images.
50%UnverifiedThere can be a vast gap between a technology's performance in a laboratory setting and its actual effectiveness in real, complex, target-sparse public environments.
50%UnverifiedIn a facial recognition system with 99.9% accuracy, if there are only 10 real targets among 1 million people, approximately 99% of the roughly 1,010 alerts generated would be false positives.
50%UnverifiedFacial data is a biometric identifier that is nearly impossible to change — passwords can be reset, but faces cannot be replaced.
50%UnverifiedAirports and Madison Square Garden in New York are using facial recognition technology.
50%UnverifiedThe greatest advantage of voiceprint recognition over facial recognition is that it requires no additional hardware and can be collected by any device with a microphone.
50%UnverifiedIn the field of face recognition, a low-discriminability feature space corresponds to a distribution where both intra-class variance and inter-class variance are small, causing Euclidean distance or cosine similarity metrics to fail.
50%UnverifiedFacial recognition systems have significantly higher misidentification rates for people with darker skin than for those with lighter skin.
50%UnverifiedAirports have in recent years introduced large numbers of technologies such as self-service check-in, facial recognition, and automated security screening.
50%UnverifiedVisual data is more privacy-sensitive than audio data; facial recognition systems can extract 128-dimensional or 512-dimensional facial feature vectors from video frames to compare against a database.
50%