Pose Invariant Face Recognition Using Principal Component Analysis
| Autor: | Akash Krishna Srivastava, Koushlendra Kumar Singh, H. Sneha, Diksha |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Individual based
ComputingMethodologies_PATTERNRECOGNITION Computational complexity theory Computer science business.industry Dimensionality reduction Principal component analysis ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Pattern recognition Artificial intelligence Invariant (mathematics) business Facial recognition system |
| Zdroj: | Advances in Intelligent Systems and Computing ISBN: 9789811540318 |
| Popis: | Facial recognition is a propitious field that has been actively studied and researched upon in the recent years.The main challenge in face recognition is the various poses exhibited by subject. It makes it difficult to accurately identify the individual based on facial recognition. The present work encompasses of a new frontalization algorithm and facial recognition using principal component analysis. The three-dimensional face geometry model has been used to produce better facial features. The Zhu-Ramanan detector has been used for detection of facial features. Principal component analysis is a statistical approach used for face recognition which involves dimensionality reduction. The proposed approach was validated on two different databases. Outcomes of the proposed approach clearly show the reduction in computational complexity of process of face recognition. The approach successfully recognizes faces up to \(\pm 70^{\circ }\) yaw. It outperforms other methods for face recognition in terms of its efficiency to recognize faces. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|