Selective Deep Convolutional Features for Image Retrieval
This work provides an incremental improvement for image search systems by enhancing feature selection and aggregation methods.
The paper tackled the problem of improving image retrieval by selecting a subset of convolutional features to reduce redundancy and address burstiness, achieving state-of-the-art retrieval accuracy.
Convolutional Neural Network (CNN) is a very powerful approach to extract discriminative local descriptors for effective image search. Recent work adopts fine-tuned strategies to further improve the discriminative power of the descriptors. Taking a different approach, in this paper, we propose a novel framework to achieve competitive retrieval performance. Firstly, we propose various masking schemes, namely SIFT-mask, SUM-mask, and MAX-mask, to select a representative subset of local convolutional features and remove a large number of redundant features. We demonstrate that this can effectively address the burstiness issue and improve retrieval accuracy. Secondly, we propose to employ recent embedding and aggregating methods to further enhance feature discriminability. Extensive experiments demonstrate that our proposed framework achieves state-of-the-art retrieval accuracy.