Adaptive Co-weighting Deep Convolutional Features For Object Retrieval
This work addresses image retrieval efficiency for computer vision applications, but it is incremental as it builds on existing aggregation methods.
The paper tackles the problem of aggregating deep convolutional features for image retrieval by proposing an unsupervised method that uses an adaptive Gaussian filter and an element-value sensitive vector to co-weight features, resulting in improved discrimination power and outperforming recent aggregation approaches by a considerable margin.
Aggregating deep convolutional features into a global image vector has attracted sustained attention in image retrieval. In this paper, we propose an efficient unsupervised aggregation method that uses an adaptive Gaussian filter and an elementvalue sensitive vector to co-weight deep features. Specifically, the Gaussian filter assigns large weights to features of region-of-interests (RoI) by adaptively determining the RoI's center, while the element-value sensitive channel vector suppresses burstiness phenomenon by assigning small weights to feature maps with large sum values of all locations. Experimental results on benchmark datasets validate the proposed two weighting schemes both effectively improve the discrimination power of image vectors. Furthermore, with the same experimental setting, our method outperforms other very recent aggregation approaches by a considerable margin.