The recent advances brought by deep learning allowed to improve the performance in image retrieval tasks. Through the many convolutional layers, available in a Convolutional Neural Network (CNN), it is possible to obtain a hierarchy of features from the evaluated image. At every step, the patches extracted are smaller than the previous levels and more representative. Following this idea, this paper introduces a new detector applied on the feature maps extracted from pre-trained CNN. Specifically, this approach lets to increase the number of features in order to increase the performance of the aggregation algorithms like the most famous and used VLAD embedding. The proposed approach is tested on different public datasets: Holidays, Oxford5k, Paris6k and UKB.
A Dense-Depth Representation for VLAD descriptors in Content-Based Image Retrieval / Magliani, Federico; Fontanini, Tomaso; Prati, Andrea. - ELETTRONICO. - 11241:(2018), pp. 662-671. (Intervento presentato al convegno 13th International Symposium on Visual Computing (ISVC) tenutosi a Las Vegas, NV (USA) nel 19-21 November 2018) [10.1007/978-3-030-03801-4_58].
A Dense-Depth Representation for VLAD descriptors in Content-Based Image Retrieval
Federico Magliani
Methodology
;Tomaso FontaniniWriting – Original Draft Preparation
;Andrea PratiSupervision
2018-01-01
Abstract
The recent advances brought by deep learning allowed to improve the performance in image retrieval tasks. Through the many convolutional layers, available in a Convolutional Neural Network (CNN), it is possible to obtain a hierarchy of features from the evaluated image. At every step, the patches extracted are smaller than the previous levels and more representative. Following this idea, this paper introduces a new detector applied on the feature maps extracted from pre-trained CNN. Specifically, this approach lets to increase the number of features in order to increase the performance of the aggregation algorithms like the most famous and used VLAD embedding. The proposed approach is tested on different public datasets: Holidays, Oxford5k, Paris6k and UKB.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.