Contrastive Self-Supervised Hashing With Dual Pseudo Agreement

Li, Yang; Wang, Yapeng; Zhang, Miao; Wang, Jiabao; Zhang, Rui

doi:10.1109/access.2020.3022672

Cited by 7 publications

(4 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The NRDH method is compared with several popular hash learning algorithms: FPH [16], HashNet [18], NSPH [26], CSH [27], DOH [28], and RODH [29]. Table 1 compares the MAP of our DNRH algorithm and existing hash learning algorithms on the CIFAR-10 dataset.…”

Section: B Experimental Resultsmentioning

confidence: 99%

“…Quantization loss reduces the error caused by binarizing realvalued feature representations to hash codes, and bit balance loss weakens hash code bias. CSH [27] proposed a simple but effective approach to self-supervised hash learning based on dual pseudo-agreement. By adding a consistency constraint, this method can prevent corrupted labels and encourage generalization for effective knowledge distillation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Non-Relaxation Deep Hashing Method for Fast Image Retrieval

2023

IEEE Access

View full text Add to dashboard Cite

Deep hashing methods utilize an end-to-end framework to mutually learn feature representations and hash codes, thereby achieving a better retrieval performance. Traditional supervised hashing methods adopt handcrafted features for hashing function learning and then generate hash codes through classification and quantization. The lack of adaptability and independence of the quantization procedure leads to low retrieval accuracy of supervised hashing methods with handcrafted features in image retrieval. In this study, a non-relaxation deep hashing method for fast image retrieval is proposed. In this method, a differentiable host thresholding function is used to encourage hash-like codes to approach -1 or 1 non-linearly at the output of the convolutional neural, instead of the symbol function for quantization used in the traditional method. The output of the host thresholding function is directly used to compute the network training error, and a loss function is elaborately designed with the norm to constrain each bit of the hash-like code to be as binary as possible. Finally, a symbol function is added outside the trained network model to generate binary hash codes for image storage and retrieval in a low-dimensional binary space. Extensive experiments on two large-scale public datasets show that our method can effectively learn image features, generate accurate binary hash codes, and outperform state-of-the-art methods in terms of the mean average precision.

show abstract

Section: B Experimental Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Non-Relaxation Deep Hashing Method for Fast Image Retrieval

2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…However, these models did not make full use of the pair information. While contrastive learning [8,16,31] aims to leverage pair information, the success of contrastive hashing in tasks like image retrieval [11,15,19] inspired numerous researchers to investigate its application in cross-modal hashing.…”

Section: Related Workmentioning

confidence: 99%

Category-Level Contrastive Learning for Unsupervised Hashing in Cross-Modal Retrieval

Xu,

Luo,

Lai

et al. 2024

Data Sci. Eng.

View full text Add to dashboard Cite

Unsupervised hashing for cross-modal retrieval has received much attention in the data mining area. Recent methods rely on image-text paired data to conduct unsupervised cross-modal hashing in batch samples. There are two main limitations for existing models: (1) learning of cross-modal representations is restricted to batches; (2) semantically similar samples may be wrongly treated as negative. In this paper, we propose a novel category-level contrastive learning for unsupervised cross-modal hashing, which alleviates the above problems and improves cross-modal query accuracy. To break the limitation of learning in small batches, a selected memory module is first proposed to take global relations into account. Then, we obtain pseudo labels through clustering and combine the labels with the Hadamard Matrix for category-centered learning. To reduce wrong negatives, we further propose a memory bank to store clusters of samples and construct negatives by selecting samples from different categories for contrastive learning. Extensive experiments show the significant superiority of our approach over the state-of-the-art models on MIRFLICKR-25K and NUS-WIDE datasets.

show abstract

“…With the rise of big data, the use and rapid development of big data in the field of deep learning and computer vision has been promoted, and the deep learning and computer vision based on big data can complete machine training more effectively and accurately. Deep learning and computer vision and big data analysis are techniques that use existing computer system models to convert a large amount of data or big data, acquired by computers into useful information [1][2][3]. The larger the scale of the data used, the better the training effect of deep learning and computer vision, the more accurate, the more content recognition, and the reduction of overfitting and underfitting phenomena.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning Distributed Architecture Design Implementation for Computer Vision

Zhang

2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

In the era of big data, to achieve an efficient deep learning and computer vision system for big data, developers need to build a computerized deep learning and computer vision system, and the system can simultaneously complete the tasks of deep learning and computer vision and large-scale data processing. The existing training dataset is reused, and the scene information is small, which cannot meet the needs of large-scale machine training, so it is necessary to include large-scale data, distributed computer system to complete the training. How to meet the training accuracy requirements of deep learning models and minimize the resource cost within the constrained time is a major challenge for distributed deep learning systems. Resource and batch size hyperparameter allocation are the main approaches to optimize the training accuracy and resource cost of models. Existing works have independently configured resources and batch size hyperparameters in terms of computational efficiency and training accuracy, respectively. However, the impact of the two types of configurations on model training accuracy and resource cost has complex dependencies, and it is difficult to achieve the goals of satisfying the model training accuracy requirements and minimizing the resource cost simultaneously by the existing independent configuration methods. To address these problems, this paper proposes a collaborative resource-batch size optimization configuration method for distributed deep learning systems. This method was firstly based on the monotonic function relationship between resource allocation and batch size hyperparameter allocation and model training time and training accuracy, and we select the order-preserving regression theoretical tool to build a model prediction model for single-round complete training time and final training accuracy for computer vision target classification and recognition, respectively; then, we use the abovementioned models together to solve the resource and batch size optimal allocation solutions to meet the model training accuracy requirements with the goal of minimizing resource cost. The optimal allocation of resources and batch size to meet the training accuracy requirements of the model is solved. In this paper, we evaluate the performance of the proposed method for computer vision target recognition based on the proposed distributed deep learning system.

show abstract

Contrastive Self-Supervised Hashing With Dual Pseudo Agreement

Cited by 7 publications

References 45 publications

Non-Relaxation Deep Hashing Method for Fast Image Retrieval

Non-Relaxation Deep Hashing Method for Fast Image Retrieval

Category-Level Contrastive Learning for Unsupervised Hashing in Cross-Modal Retrieval

Deep Learning Distributed Architecture Design Implementation for Computer Vision

Contact Info

Product

Resources

About