Deep joint image filtering

Yijun Li, Jia Bin Huang, Narendra Ahuja, Ming Hsuan Yang

Research output: Chapter in Book/Report/Conference proceedingConference contribution


Joint image filters can leverage the guidance image as a prior and transfer the structural details from the guidance image to the target image for suppressing noise or enhancing spatial resolution. Existing methods rely on various kinds of explicit filter construction or handdesigned objective functions. It is thus difficult to understand, improve, and accelerate them in a coherent framework. In this paper, we propose a learning-based approach to construct a joint filter based on Convolutional Neural Networks. In contrast to existing methods that consider only the guidance image, our method can selectively transfer salient structures that are consistent in both guidance and target images. We show that the model trained on a certain type of data, e.g., RGB and depth images, generalizes well for other modalities, e.g., Flash/Non-Flash and RGB/NIR images. We validate the effectiveness of the proposed joint filter through extensive comparisons with state-of-the-art methods.

Original languageEnglish (US)
Title of host publicationComputer Vision - 14th European Conference, ECCV 2016, Proceedings
EditorsBastian Leibe, Jiri Matas, Nicu Sebe, Max Welling
Number of pages16
ISBN (Print)9783319464923
StatePublished - 2016
Event14th European Conference on Computer Vision, ECCV 2016 - Amsterdam, Netherlands
Duration: Oct 11 2016Oct 14 2016

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9908 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Other14th European Conference on Computer Vision, ECCV 2016


  • Deep convolutional neural networks
  • Joint filtering

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Deep joint image filtering'. Together they form a unique fingerprint.

Cite this