Hierarchical bilinear convolutional neural network for image classification

Zhang, Xiang; Tang, Lei; Luo, Hangzai; Guan, Ziyu; Chen, Long; Zhao, Chao; Peng, Jinye; Fan, Jianping

doi:10.1049/cvi2.12023

Cited by 10 publications

(12 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Zhang et al [36] developed Hierarchical Bilinear CNN model by combining convolutional networks with multitask learning on the hierarchical visual structures. Hierarchical bilinear CNN utilized VGG16 for constructing a network of inner output branches.…”

Section: Deep Learning Based Image Classification Modelsmentioning

confidence: 99%

An abbreviated review of deep learning-based image classification models

Abbood

Al-Turfi

Almajmaie

2023

IJEECS

View full text Add to dashboard Cite

Image classification is an extensively researched sub-fields of computer vision implemented in face recognition, self-driving, medical image segmentation, biological identification, and others. Traditional models of image classification require manual construction of feature extraction techniques and classification accuracy which are closely associated with these utilized techniques. During the rapid progress of multimedia technologies, the number of images that require classification got bigger, and this led to making image classification more complicated, hence, the manual construction of feature extraction techniques consumes more time and provides lower accuracy. In the recent decade, deep learning-based models have appeared in various applications. These models hold the merits of an effective extraction of image features, low-weight features filtering, a large capacity for processing, and higher classification speed and accuracy. Thus, lots of researchers have attempted to utilize deep learning algorithms, especially convolutional neural networks (CNNs) for image classification. Therefore, this paper concentrates on providing an abbreviated review of deep learning-based image classification models, by covering the recently utilized deep learning algorithms, comparing various related works and benchmark datasets mentioned in this paper, and summarizing the fundamental analysis and discussion.

show abstract

Section: Deep Learning Based Image Classification Modelsmentioning

confidence: 99%

An abbreviated review of deep learning-based image classification models

Abbood

Al-Turfi

Almajmaie

2023

IJEECS

View full text Add to dashboard Cite

show abstract

“…To that moment, branches were treated independently, but authors such as Inoue et al [16] and Zhang et al [17] demonstrated the benefit of interconnecting branches. Indeed, they showed that branches can complement information from different hierarchy levels during training.…”

Section: Related Workmentioning

confidence: 99%

“…In [16], branches are connected similary to [17], but instead of connecting the representation extracted from the central block, they connect the pre-softmax layer y…”

Section: A Branched Architecturementioning

confidence: 99%

“…}, which have been annotated for that level of the class hierarchy. Previous works (see, for example, [15] and [17]) often assumed that data were classified at a finer level in the hierarchy. In this case, level-wise annotations can be obtained by tracing the hierarchy back.…”

Section: Learningmentioning

confidence: 99%

“…This way, the prediction for a given level is explicitly conditioned on the previous one. More recently, [17] proposed to connect a branch with all the previous branches in the network. Finally, in [18], the authors used a similar connectivity pattern, but instead of propagating hidden representations, they propagate probability estimates from each branch.…”

mentioning

confidence: 99%

See 2 more Smart Citations

An Attention-Based Architecture for Hierarchical Classification With CNNs

2023

View full text Add to dashboard Cite

Branch Convolutional Neural Nets have become a popular approach for hierarchical classification in computer vision and other areas. Unfortunately, these models often led to hierarchical inconsistency: predictions for the different hierarchy levels do not necessarily respect the class-subclass constraints imposed by the hierarchy. Several architectures to connect the branches have arisen to overcome this limitation. In this paper, we propose a more straightforward and flexible method: let the neural net decide how these branches must be connected. We achieve this by formulating an attention mechanism that dynamically determines how branches influence each other during training and inference. Experiments on image classification benchmarks show that the proposed method can outperform stateof-the-art models in terms of hierarchical performance metrics and consistency. Furthermore, although sometimes we found a slightly lower performance at the deeper level of the hierarchy, the model predicts much more accurately the ground-truth path between a concept and its ancestors in the hierarchy. This result suggests that the model does learn not only local class memberships but also hierarchical dependencies between concepts.

show abstract

Clean, performance‐robust, and performance‐sensitive historical information based adversarial self‐distillation

Li,

Hu,

Huo

et al. 2024

IET Computer Vision

View full text Add to dashboard Cite

Adversarial training suffers from poor effectiveness due to the challenging optimisation of loss with hard labels. To address this issue, adversarial distillation has emerged as a potential solution, encouraging target models to mimic the output of the teachers. However, reliance on pre‐training teachers leads to additional training costs and raises concerns about the reliability of their knowledge. Furthermore, existing methods fail to consider the significant differences in unconfident samples between early and late stages, potentially resulting in robust overfitting. An adversarial defence method named Clean, Performance‐robust, and Performance‐sensitive Historical Information based Adversarial Self‐Distillation (CPr & PsHI‐ASD) is presented. Firstly, an adversarial self‐distillation replacement method based on clean, performance‐robust, and performance‐sensitive historical information is developed to eliminate pre‐training costs and enhance guidance reliability for the target model. Secondly, adversarial self‐distillation algorithms that leverage knowledge distilled from the previous iteration are introduced to facilitate the self‐distillation of adversarial knowledge and mitigate the problem of robust overfitting. Experiments are conducted to evaluate the performance of the proposed method on CIFAR‐10, CIFAR‐100, and Tiny‐ImageNet datasets. The results demonstrate that the CPr&PsHI‐ASD method is more effective than existing adversarial distillation methods in enhancing adversarial robustness and mitigating robust overfitting issues against various adversarial attacks.

show abstract

Hierarchical bilinear convolutional neural network for image classification

Cited by 10 publications

References 28 publications

An abbreviated review of deep learning-based image classification models

An abbreviated review of deep learning-based image classification models

An Attention-Based Architecture for Hierarchical Classification With CNNs

Clean, performance‐robust, and performance‐sensitive historical information based adversarial self‐distillation

Contact Info

Product

Resources

About