دانلود رایگان مقاله انگلیسی چالش های مرتبط با یادگیری عمیق در تشخیص چهره بدون محدودیت - IEEE 2018

Abstract

Recently deep learning has become dominant in face recognition and many other artificial intelligence areas. We raise a question: Can deep learning truly solve the face recognition problem? If not, what is the challenge for deep learning methods in face recognition? We think that the face image quality issue might be one of the challenges for deep learning, especially in unconstrained face recognition. To investigate the problem, we partition face images into different qualities, and evaluate the recognition performance, using the state-of-the-art deep networks. Some interesting results are obtained, and our studies can show directions to promote the deep learning methods towards high-accuracy and practical use in solving the hard problem of unconstrained face recognition.

CONCLUSION

We have proposed to partition face images based on quality for investigating critical issues in unconstrained face recognition. Based on quality partition, we have developed FR protocols for cross-quality face identification and verification on two public databases. Some representative deep learning methods have been evaluated under our settings for unconstrained FR. We have shown that the face image quality variations are a grand challenge for deep learning in performing unconstrained FR, even though a variety of face images have been fed into the training of deep networks. Our study suggests the direction to promote deep learning techniques towards high-accuracy recognition in practice.