In this work, we present the novel Inter-GIP Distances (IGD) feature and its integration into the Gestalt Interest Points (GIP) image descriptor. With the ongoing growth of visual data, efficient image descriptor methods are becoming more and more important. Several local point-based description methods have been defined in the past decades. Accuracy and descriptor size are important factors when selecting the appropriate method for a given retrieval problem. The method presented in this work describes images with only a few very compact descriptors. To test our descriptor, we developed an image classification prototype and conducted several experiments with a publicly available horses dataset and a food dataset. Our experiments show that only a few of the very compact GIP image descriptors are necessary to quickly classify the images from the datasets with high accuracy. Furthermore, we compared our experimental results to state-of-the-art local point-based description methods and found that our method is highly competitive.
M. Hörhan, H. Eidenberger: "The Gestalt Interest Points Distance Feature for Compact and Accurate Image Description"; Talk: IEEE International Symposium on Signal Processing and Information Technology, Bilbao; 12-18-2017 - 12-20-2017; in: "Proceedings IEEE International Symposium on Signal Processing and Information Technology", (2017), 5 pages.
Click into the text area and press Ctrl+A/Ctrl+C or ⌘+A/⌘+C to copy the BibTeX into your clipboard… or download the BibTeX.