CFP: Special Issue on Deep Learning in Image and Video Retrieval

in the International Journal of Multimedia Information Retrieval

(published by Springer with support from ACM SIGMM and the MIR Society)


Deep learning has led to numerous major advances in content based retrieval, multimedia analysis and computer vision. The new approaches have been shown to reach near-human levels of semantic understanding of visual media by using deep neural networks which simultaneously learn both features and the classifier from the training data. Deep learning has also become a powerful tool to produce new features and representations such as semantic segmentations and generative adversarial network (GAN) images which can be used for improving content based retrieval systems. All of these advances have also shown the importance of both new deep architectures (VGG, ResNet, GANs, etc.) and high quality training data sets such as ImageNet, NUS-Wide and MIRFLICKR-1million.

This special issue aims to capture the state-of-the-art in deep learning in the context of image and video retrieval. We are especially interested in novel deep architectures for content based retrieval, original high quality benchmark datasets for deep learning, and new insights into deep learning systems both theoretical and empirical, especially ones which examine and critically assess the state-of-the-art


Due to numerous requests there is an extension until April 29, 2019

About the Journal

The International Journal of Multimedia Information Retrieval is indexed by Scopus (2016 impact factor of 2.2) and Web of Science. For more information see and

For more details (including topics of interest) please see PDF Call for Papers or JPG Call for Papers



Guest Editors

Ard Oerlemans, Google, USA,

Yanming Guo, NUDT, China,

Michael Lew, Leiden University, Netherlands,

Tat-Seng Chua, NUS, Singapore,