Unsupervised Deep Hashing with Similarity-Adaptive and Discrete Optimization
By Fumin Shen, Yan Xu, Li Liu, Yang Yang, Zi-Liang Huang, Heng Tao Shen. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018
372 citations recorded by the source corpus.
Recent vision and learning studies show that learning compact hash codes can facilitate massive data processing with significantly reduced storage and computation. Particularly, learning deep hash functions has greatly improved the retrieval performance, typically under the semantic supervision. In contrast, current unsupervised deep hashing algorithms can hardly achieve satisfactory performance due to either the relaxed optimization or absence of similarity-sensitive objective. In this work, we propose a simple yet effective unsupervised hashing framework, named Similarity-Adaptive Deep Hashing (SADH), which alternatingly proceeds over three training modules: deep hash model training, similarity graph updating and binary code optimization. The key difference from the widely-used two-step hashing method is that the output representations of the learned deep model help update the similarity graph matrix, which is then used to improve the subsequent code optimization. In addition, for producing high-quality binary codes, we devise an effective discrete optimization algorithm which can directly handle the binary constraints with a general hashing loss. Extensive experiments validate the efficacy of SADH, which consistently outperforms the state-of-the-arts by large gaps.
Subfields: Computer Vision and Pattern Recognition
View Paper: https://www.semanticscholar.org/paper/38cb2bab8bb1f9ec6a52b0e049536467408cc081
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Vision and Pattern Recognition
Citations in the corpus
- Structured AutoEncoders for Subspace Clustering (2018, background): “vated by the huge success achieved by deep representation learning [25]–[29], we propose a new deep subspace clus-”
- Deep Multi-View Enhancement Hashing for Image Retrieval (2020, methodology): “Compared with unsupervised hashing algorithm [35] such as LSH [36], KLSH [37]”
- Binary Multi-View Clustering (2019, background): “Fortunately, recent advancement of binary code learning [27], [28], [29] provides a preferable solution to overcome the above challenges due to its efficiency in computation and space.”
- Learning Cross-Modal Common Representations by Private–Shared Subspaces Separation (2020, background): “Traditional unimodal retrieval applications, such as document retrieval [1], [2] and image/video search [3]–[6], limitedly concentrate on finding the retrieval results for the query data within the same modality.”
- A Decade Survey of Content Based Image Retrieval Using Deep Learning (2020, methodology): “[107] by employing the training of the deep hash model, updating the similarity graph and optimizing the binary codes.”
- Two-Stream Deep Hashing With Class-Specific Centers for Supervised Image Search (2020, background): “Deep learning-based hashing methods [43], [51]–[65] have also been proposed, showing that deep networks can help to facilitate more effective hash code learning.”
- Zero-Shot Sketch-Image Hashing (2018, methodology): “As an extension of conventional data hashing techniques [51, 20, 39, 54, 53], cross-modal hashing [4, 13, 72, 36, 33, 26, 5, 6] show great potential in retrieving heterogeneous data with high efficiency due to the computationally cheap Hamming space matching, which is recently adopted to large-scale”
- The Gap of Semantic Parsing: A Survey on Automatic Math Word Problem Solvers (2018, background): “nificantly boost the accuracy. For instance, SIFT [71], [72] and other local descriptors [73], [74], [75] have been intensively used in the domain of object recognition and image retrieval [76], [77], [78] for decades as they are invariant to scaling, orientation and illumination changes. Consequentl”