Abstract
Interested face regions have the low-resolution problem in the video surveillance because the distance between face and camera is far. Thus, the high-resolution (HR) faces need to be reconstructed from low-resolution (LR) faces for further processing. Typical face hallucination based on patch-wise sparse coding can achieve better results but have very high complexity for training. In order to reduce the complexity, this paper proposes a method which uses K-means++ clustering instead of sparse coding to obtain an over-complete dictionary pair. Then, the least angle regression (LARS) algorithm is utilized to calculate the coefficients and reconstruct the high-resolution faces. The experimental results show that proposed algorithm can effectively reduce complexity in condition of irregular LR faces. In addition, the comparisons also prove that the proposed method can improve the value of PSNR and SSIM in the same database.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
