Joost van de Weijer
Joost van de Weijer is a computer vision researcher at the Computer Vision Center in Barcelona, where he leads the LAMP team. He completed a PhD in physics-based computer vision at the University of Amsterdam in 2005, held a Marie Curie fellowship at INRIA from 2005 to 2007, and a Ramón y Cajal fellowship at UAB from 2008 to 2012. His 2007 color-constancy framework combined Grey-World, White-Patch, and Grey-Edge methods; in 2023, he coauthored a survey of class-incremental image classification. In July 2026, he co-chaired a CVC workshop session on generative models and graph neural networks; as of October 2026, he is an ELLIS Fellow.
- Works
- 307
- Citations
- 19,010
- h-index
- 64
Publication numbers from OpenAlex, as of 15 October 2026.
Timeline
-
2005
A doctorate in computer vision
At the University of Amsterdam, he completes a PhD in 2005 focused on physics-based computer vision for improved color understanding. His thesis research develops object-color descriptions that remain stable under shadow changes, viewpoint variation, object position, and illuminant color.
-
2005–2007
A Marie Curie fellowship
From 2005 to 2007, he holds a Marie Curie Intra-European fellowship with the LEAR team at INRIA Rhône-Alpes and applies his color research to object recognition.
-
2007
A color-constancy framework
His 2007 paper with Theo Gevers and Arjan Gijsenij combines Grey-World, White-Patch, and Grey-Edge methods in one color-constancy framework.
-
2008–2012
A Ramón y Cajal fellowship
From 2008 to 2012, he holds a Ramón y Cajal fellowship at the Universitat Autònoma de Barcelona.
-
2023
A class-incremental survey
In 2023, he coauthors a survey of class-incremental learning for image classification that reviews existing methods and evaluates thirteen approaches.
-
As of October 2026
Leading LAMP at CVC
He leads the Learning and Machine Perception team at the Computer Vision Center and is a Fellow of the ELLIS Society. The team studies continual learning and machine learning for semantic understanding of visual data. In July 2026, he co-chairs a CVC workshop session on generative models and graph neural networks.
Sources and credits
Photo: We have no published photo of Joost van de Weijer yet. Send us a free-licensed photo.
- Joost van de Weijer
- Photometric Robust Features for Object Recognition in Colour Images
- Going from task to class-incremental learning
- Source-code | Color Constancy
- Class-Incremental Learning: Survey and Performance Evaluation on Image Classification
- Joost van de Weijer | European Laboratory for Learning and Intelligent Systems
- Learning and Machine Perception - CVC Computer Vision Center
- CVCRD 2026 – 21st CVC Workshop on Computer Vision Trends and Challenges