Deblurring Face Images Using Uncertainty Guided Multi-Stream Semantic Networks

You are here

Top Reasons to Join SPS Today!

1. IEEE Signal Processing Magazine
2. Signal Processing Digital Library*
3. Inside Signal Processing Newsletter
4. SPS Resource Center
5. Career advancement & recognition
6. Discounts on conferences and publications
7. Professional networking
8. Communities for students, young professionals, and women
9. Volunteer opportunities
10. Coming soon! PDH/CEU credits
Click here to learn more.

Deblurring Face Images Using Uncertainty Guided Multi-Stream Semantic Networks

By: 
Rajeev Yasarla; Federico Perazzi; Vishal M. Patel

We propose a novel multi-stream architecture and training methodology that exploits semantic labels for facial image deblurring. The proposed Uncertainty Guided Multi-Stream Semantic Network (UMSN) processes regions belonging to each semantic class independently and learns to combine their outputs into the final deblurred result. Pixel-wise semantic labels are obtained using a segmentation network. A predicted confidence measure is used during training to guide the network towards the challenging regions of the human face such as the eyes and nose. The entire network is trained in an end-to-end fashion. Comprehensive experiments on three different face datasets demonstrate that the proposed method achieves significant improvements over the recent state-of-the-art face deblurring methods. View Code.

SPS on Twitter

  • Attention Region 7! Join us today at 10:00 AM ET for the Region 7 Member Forum, during which SPS leaders and staff… https://t.co/plFMa08CcD
  • RESCHEDULED WEBINAR: "Hybrid Digital and Analog Beamforming Design for Large-Scale Antenna Arrays," originally sche… https://t.co/AtbofklFxP
  • Starting soon: The 2020 IEEE TENCON is happening now, and Women in Signal Processing leaders will be presenting in… https://t.co/60fQiVSanR
  • THIS FRIDAY: The Brain Space Initiative Talk Series continues with Daniele Marinazzo presenting "Connectivity throu… https://t.co/GwIl1SNNmq
  • The SPACE Webinar Series continues this Tuesday, 17 November at 11:00 AM ET when Yoram Bresler presents "Two Topics… https://t.co/FZs8Z9GanT

SPS Videos


Signal Processing in Home Assistants

 


Multimedia Forensics


Careers in Signal Processing             

 


Under the Radar