Co-occurrence of deep convolutional features for image search
This work addresses image retrieval for computer vision applications, presenting an incremental improvement over existing methods.
The paper tackles image search by proposing a new co-occurrence representation of deep convolutional features to enhance image descriptors, achieving improved performance on well-known image retrieval datasets.
Image search can be tackled using deep features from pre-trained Convolutional Neural Networks (CNN). The feature map from the last convolutional layer of a CNN encodes descriptive information from which a discriminative global descriptor can be obtained. We propose a new representation of co-occurrences from deep convolutional features to extract additional relevant information from this last convolutional layer. Combining this co-occurrence map with the feature map, we achieve an improved image representation. We present two different methods to get the co-occurrence representation, the first one based on direct aggregation of activations, and the second one, based on a trainable co-occurrence representation. The image descriptors derived from our methodology improve the performance in very well-known image retrieval datasets as we prove in the experiments.