Glasses Removal from Facial Image Using Recursive PCA Reconstruction
This paper proposes a new method of removing glasses from human frontal face images. We first detect regions occluded by glasses. Then we generate a naturally looking glassless facial image by recursive PCA reconstruction. The resulting image has no trace of glass frame, nor the reflection and shade made by the glasses. The experimental results show that the proposed method is an effective solution to the problem of glass occlusion, and, we believe, it can be applied to enhancing the performance of face recognition systems.
Unable to display preview. Download preview PDF.
- Chellappa, R., Wilson, C. L., Sirohey, S.: Human and Machine Recognition of Faces: A Survey. Proc. of IEEE, Vol. 83, No. 5 (May 1995) 705–740Google Scholar
- Jing, Z., Mariani, R.: Glasses Detection and Extraction by Deformable Contour. Proc. of Int’l Conf. on Pattern Recognition, Barcelona, Spain (2000) 933–936Google Scholar
- Saito, Y., Kenmochi, Y., Kotani, K.: Estimation of eyeglassless facial images using principal component analysis. Proc. of Int’l Conf. on Image Processing, Vol. 4, Kobe, Japan (Oct. 1999) 197–201Google Scholar
- Wu H. et al.: Glasses Frame Detection with 3D Hough Transform. Proc. of 16th Int. Conf. on Pattern Recognition, Vol. II, Quebec, Canada (Aug. 2002) 346–349Google Scholar
- Kim, J.K. et al.: Automatic FDP/FAP generation from an image sequence. Proc. of IEEE Int’l Symposium of Circuits and Systems, Geneva, Switzerland (May 2000) 40–43Google Scholar