TR2018-140

Super-resolution of Very Low-Resolution Faces from Videos


    •  Ataer-Cansizoglu, E., Jones, M.J., "Super-resolution of Very Low-Resolution Faces from Videos", British Machine Vision Conference (BMVC), September 2018.
      BibTeX TR2018-140 PDF
      • @inproceedings{Ataer-Cansizoglu2018sep,
      • author = {Ataer-Cansizoglu, Esra and Jones, Michael J.},
      • title = {Super-resolution of Very Low-Resolution Faces from Videos},
      • booktitle = {British Machine Vision Conference (BMVC)},
      • year = 2018,
      • month = sep,
      • url = {https://www.merl.com/publications/TR2018-140}
      • }
  • MERL Contact:
  • Research Areas:

    Artificial Intelligence, Computer Vision, Machine Learning

Faces appear in low-resolution video sequences in various domains such as surveillance. The information accumulated over multiple frames can help super-resolution for high magnification factors. We present a method to super-resolve a face image using the consecutive frames of the face in the same sequence. Our method is based on a novel multi-input-single-output framework with a Siamese deep network architecture that fuses multiple frames into a single face image. Contrary to existing work on video super-resolution, it is model free and does not depend on facial landmark detection that might be difficult to handle for very low-resolution faces. The experiments show that the use of multiple frames as input improves the performance compared to single-inputsingle-output
systems.