Abstract | ||
---|---|---|
•We propose a simple yet effective quantization to embed deep binary codes.•Activations from multiple CNN layers function together through weighted score fusion in the proposed framework.•Handcrafted local descriptor SIFT, as a kind of low level feature, can also be combined in our fusion procedure.•Regularized diffusion process are customized on the ranking list to make the similarity estimation vary smoothly.•Extensive experiments are conducted on four public datasets, and state-of-the-art results are obtained on Holidays and UKBench datasets. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1016/j.neucom.2018.06.014 | Neurocomputing |
Keywords | Field | DocType |
Image retrieval,Bag-of-Words,Feature fusion,Diffusion process,Re-ranking | Inverted index,Scale-invariant feature transform,Feature vector,Pattern recognition,Convolutional neural network,Binary code,Image retrieval,Artificial intelligence,Contextual image classification,Mathematics,Binary number | Journal |
Volume | ISSN | Citations |
314 | 0925-2312 | 3 |
PageRank | References | Authors |
0.37 | 35 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ying Li | 1 | 11 | 1.82 |
Xiang-Wei Kong | 2 | 212 | 15.09 |
Haiyan Fu | 3 | 10 | 2.85 |
Qi Tian | 4 | 6443 | 331.75 |