A nuclear-norm model for multi-frame super-resolution reconstruction from video clips

Rui ZHAO, Raymond HF CHAN*

*Corresponding author for this work

Research output: Book Chapters | Papers in Conference ProceedingsBook ChapterResearchpeer-review


We propose a variational approach to obtain super-resolution images from multiple low-resolution frames extracted from video clips. First the displacement between the low-resolution frames and the reference frame is computed by an optical flow algorithm. Then a low-rank model is used to construct the reference frame in high resolution by incorporating the information of the low-resolution frames. The model has two terms: a 2-norm data fidelity term and a nuclear-norm regularization term. Alternating direction method of multipliers is used to solve the model. Comparison of our methods with other models on synthetic and real video clips shows that our resulting images are more accurate with less artifacts. It also provides much finer and discernable details.

Original languageEnglish
Title of host publicationStructured Matrices in Numerical Linear Algebra: Analysis, Algorithms and Applications
EditorsDario Andrea BINI, Fabio DI BENEDETTO, Eugene TYRTYSHNIKOV, Marc Van BAREL
PublisherSpringer, Cham
Number of pages20
ISBN (Electronic)9783030040888
ISBN (Print)9783030040871
Publication statusPublished - 2019
Externally publishedYes

Publication series

NameSpringer INdAM Series
ISSN (Print)2281-518X
ISSN (Electronic)2281-5198

Bibliographical note

Publisher Copyright:
© Springer Nature Switzerland AG 2019.


  • Image processing
  • Low-rank approximation
  • Super-resolution


Dive into the research topics of 'A nuclear-norm model for multi-frame super-resolution reconstruction from video clips'. Together they form a unique fingerprint.

Cite this