Color Constancy Model Optimization with Small Dataset via Pruning of CNN Filters

Sahar Husseini, Pouria Babahajiani, Moncef Gabbouj

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review


Color constancy is an essential part of the Image Signal Processor (ISP) pipeline, which removes the color bias of the captured image generated by scene illumination. Recently, several supervised algorithms, including Convolutional Neural Networks (CNN)-based methods, have been proved to work correctly on this problem. It is time-consuming and costly to collect many raw images of various scenes with different lighting conditions and measure corresponding illumination values. To reduce the dependence on a large scale labeled dataset and take advantage of standard CNNs architectures, we proposed an approach to create an efficient color constancy algorithm. Firstly, we utilized a structure channel pruning method to thin our baseline model. We iteratively pruned 75% channels of a specific Mobilenet version used as our model's backbone, trained on a large-scale classification dataset. It means the backbone with the classification head is used to deal with our network pruning task. Then the resulted compact model was transferred and trained on a small dataset doing color constancy. During training on the color constancy task, we applied the DSD technique. The proposed method reaches comparative performance with other state-of-The-Art models, produces fewer MACs, and can significantly decrease computational costs.

Original languageEnglish
Title of host publicationProceedings of the 2021 9th European Workshop on Visual Information Processing, EUVIP 2021
EditorsA. Beghdadi, F. Alaya Cheikh, J.M.R.S. Tavares, A. Mokraoui, G. Valenzise, L. Oudre, M.A. Qureshi
Number of pages6
ISBN (Electronic)9781665432306
ISBN (Print)9781665432313
Publication statusPublished - 2021
Publication typeA4 Article in conference proceedings
EventEuropean Workshop on Visual Information Processing - Paris, France
Duration: 23 Jun 202125 Jun 2021

Publication series

NameEuropean Workshop on Visual Information Processing
ISSN (Print)2164-974X
ISSN (Electronic)2471-8963


ConferenceEuropean Workshop on Visual Information Processing


  • Color constancy
  • network optimization
  • structured and unstructured pruning
  • transfer learning

Publication forum classification

  • Publication forum level 1

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Information Systems
  • Signal Processing


Dive into the research topics of 'Color Constancy Model Optimization with Small Dataset via Pruning of CNN Filters'. Together they form a unique fingerprint.

Cite this