Learning Compact Binary Descriptors with Unsupervised Deep Neural Networks

By Kevin Qinghong Lin, Jiwen Lu, Chu-Song Chen, Jie Zhou. Computer Vision and Pattern Recognition, 2016

348 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/7d76e1adcd1dd5947ea3fff001c9a39126a23a48

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

1. Shiv Ram Dubey
2. Junjun Jiang
3. Xingyu Jiang
4. Tongliang Liu
5. Dacheng Tao
6. Ling Shao
7. Fumin Shen
8. Yuming Shen

In Computer Vision and Pattern Recognition

1. Shiv Ram Dubey
2. Dacheng Tao
3. Cheng Deng
4. Li Liu
5. Heng Tao Shen
6. Ling Shao
7. Jingkuan Song
8. Jiayi Ma

Citations in the corpus

• Unsupervised Deep Hashing with Similarity-Adaptive and Discrete Optimization (2018, methodology): “These unsupervised models are trained by minimizing either the quantization loss [23], [25] or data reconstruction error [6], which fail to explicitly preserve the data similarities in the learned binary codes and thus achieve far below satisfactory performance.”
• Unsupervised Semantic-Preserving Adversarial Hashing for Image Search (2019, methodology): “In addition to these shallow approaches, some unsupervised deep hashing methods have also been proposed [44], [45], [47].”
• Deep Sketch Hashing: Fast Free-Hand Sketch-Based Image Retrieval (2017, methodology): “Related Work Hashing techniques [42, 16, 34, 62, 39, 30, 55, 35, 17, 14, 74, 36, 14, 70, 37, 47, 38, 54, 25, 33, 41] have recently been successfully applied to encode highdimensional features into compact similarity-preserving binary codes, which enables extremely fast similarity search by the use o”
• A Decade Survey of Content Based Image Retrieval Using Deep Learning (2020, methodology): “(2016) [103] have utilized the minimal quantization loss, evenly distributed codes and uncorrelated bits properties to generate the discriminative and compact code for image retrieval.”
• Image Matching from Handcrafted to Deep Features: A Survey (2020, methodology): “In comparison, other methods focused on binary descriptor learning are proposed in Erin Liong et al. (2015), Lin et al. (2016a) and Duan et al. (2017).”
• Two-Stream Deep Hashing With Class-Specific Centers for Supervised Image Search (2020, background): “DeepBit [67] considers original images and their corresponding rotated images as similar pairs and tries to learn hash codes to preserve this similarity.”
• Binary Generative Adversarial Networks for Image Retrieval (2017, methodology): “For example, Lin et al. [28] proposed an unsupervised deep learning approach, DeepBit, imposing three criteria on binary codes (i.e., minimal loss quantization, evenly distributed codes and uncorrelated bits) to learn a compact binary descriptor for efficient visual object matching.”