Please use this identifier to cite or link to this item:
|Scopus||Web of Science®||Altmetric|
|Title:||Mid-level deep pattern mining|
Van Den Hengel, A.
|Citation:||Proceedings / CVPR, IEEE Computer Society Conference on Computer Vision and Pattern Recognition. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2015, vol.07-12-June-2015, pp.971-980|
|Series/Report no.:||IEEE Conference on Computer Vision and Pattern Recognition|
|Conference Name:||2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2015) (7 Jun 2015 - 12 Jun 2015 : Boston, MA)|
|Yao Li, Lingqiao Liu, Chunhua Shen, Anton van den Hengel|
|Abstract:||Mid-level visual element discovery aims to find clusters of image patches that are both representative and discriminative. In this work, we study this problem from the prospective of pattern mining while relying on the recently popularized Convolutional Neural Networks (CNNs). Specifically, we find that for an image patch, activation extracted from the first fully-connected layer of a CNN have two appealing properties which enable its seamless integration with pattern mining. Patterns are then discovered from a large number of CNN activations of image patches through the wellknown association rule mining. When we retrieve and visualize image patches with the same pattern (See Fig. 1), surprisingly, they are not only visually similar but also semantically consistent. We apply our approach to scene and object classification tasks, and demonstrate that our approach outperforms all previous works on mid-level visual element discovery by a sizeable margin with far fewer elements being used. Our approach also outperforms or matches recent works using CNN for these tasks. Source code of the complete system is available online.|
|Keywords:||Visualization, detectors, association rules, training, feature extraction, accuracy|
|Rights:||© 2015 IEEE|
|Appears in Collections:||Aurora harvest 3|
Computer Science publications
Files in This Item:
|Restricted Access||967.71 kB||Adobe PDF||View/Open|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.