Technical Name Occlusion resistant face detectionrecognition system
Project Operator Tajen University/National Cheng Kung University
Project Host 王駿發
Summary The system, “Occluded Resistant Face DetectionRecognition System”, contains different scale detectors for calculating face locations. Additive angular margin loss is added into the training phase for achieving high efficiencyaccuracy. The system can achieves 80 accuracy under the 50 face occluded.
Market Potential Analysis 可抗遮蔽人臉辨識系統可以應用在門禁安全或者安防監控等相關領域。可以搭配智慧家庭的機器人或者各式各樣電子用品等應用,適用的範圍與彈性非常大。目前在人臉辨識取代傳統身分驗證的趨勢下,透過臉辨識系統取代人力,不僅提高準確率也可以減少人力資源,綜合以上應用與趨勢,此系統的開發與商品化都非常具有市場潛力。
Industrial Applicability The technology is applicable to related fields, such as security monitoringaccess security. The model training adopts two-stage training, which can quickly train the database, to make the system more convenientflexible to use.
Keyword Face detection Face recognition Face feature extraction Partial Face recognition Occlusion face recognition Face comparison Face verification Face searching Convolutional neural network Deep learning
  • Contact