Abstract | ||
---|---|---|
ABSTRACTHashing has become increasingly important for large-scale image retrieval, of which the low storage cost and fast searching are two key properties. However, existing methods adopt large neural networks, which are hard to be deployed in resource-limited devices due to the unacceptable memory and runtime overhead. We address that this huge overhead of neural networks somewhatviolates the appealing properties of hashing. In this paper, we propose a novel deep hashing method, called Binary Neural Network Hashing (BNNH) for fast image retrieval. Specifically, we construct an efficient binarized network architecture to provide lighter model and faster inference, which directly generates binary outputs as the desired hash codes without introducing the quantization loss. Besides, in order to circumvent the huge performance degradation caused by the extremely quantized activations, we introduce a simple yet effective activation-aware loss to explicitly guide the updating of activations in intermediate layers. Extensive experiments conducted on three benchmarks show that the proposed method outperforms the state-of-the-art binarization methods by large margins and validate the efficiency of BNNH. |
Year | DOI | Venue |
---|---|---|
2021 | 10.1145/3404835.3462896 | Research and Development in Information Retrieval |
Keywords | DocType | Citations |
deep hashing, image retrieval, binary neural networks | Conference | 0 |
PageRank | References | Authors |
0.34 | 0 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Wanqian Zhang | 1 | 5 | 4.11 |
Dayan Wu | 2 | 15 | 7.33 |
Yu Zhou | 3 | 98 | 22.73 |
Bo Li | 4 | 26 | 10.93 |
Wang Weiping | 5 | 335 | 63.84 |
Dan Meng | 6 | 476 | 67.10 |