Fine-Grained Visual Classification via Progressive Multi-Granularity Training of Jigsaw Patches

Du, Ruoyi; Chang, Dongliang; Bhunia, Ayan Kumar; Xie, Jiyang; Ma, Zhanyu; Song, Yi-Zhe; Guo, Jun

doi:10.48550/arxiv.2003.03836

Cited by 6 publications

(5 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For ImageNet (Deng et al 2009), ResNet-50 and ResNet-200 (He et al 2016) are adopted as target models, which are trained from scratch. For fine-grained image classification, we follow the previous works (Du et al 2020;Chen et al 2019b) and utilize pre-trained ResNet-50 and ResNet-101 models as the target network. Unless specified otherwise, the input image size is 32×32 for CIFAR while 224×224 for ImageNet, CUB-200-2011 and Stanford Dogs.…”

Section: Experiments and Resultsmentioning

confidence: 99%

SelectAugment: Hierarchical Deterministic Sample Selection for Data Augmentation

Lin

Zhang

et al. 2023

AAAI

View full text Add to dashboard Cite

Data augmentation (DA) has been extensively studied to facilitate model optimization in many tasks. Prior DA works focus on designing augmentation operations themselves, while leaving selecting suitable samples for augmentation out of consideration. This might incur visual ambiguities and further induce training biases. In this paper, we propose an effective approach, dubbed SelectAugment, to select samples for augmentation in a deterministic and online manner based on the sample contents and the network training status. To facilitate the policy learning, in each batch, we exploit the hierarchy of this task by first determining the augmentation ratio and then deciding whether to augment each training sample under this ratio. We model this process as two-step decision-making and adopt Hierarchical Reinforcement Learning (HRL) to learn the selection policy. In this way, the negative effects of the randomness in selecting samples to augment can be effectively alleviated and the effectiveness of DA is improved. Extensive experiments demonstrate that our proposed SelectAugment significantly improves various off-the-shelf DA methods on image classification and fine-grained image recognition.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

SelectAugment: Hierarchical Deterministic Sample Selection for Data Augmentation

Lin

Zhang

et al. 2023

AAAI

View full text Add to dashboard Cite

show abstract

“…Backbone IDRiD DeepDR B-CNN (ICCV15) [18] VGG16 0.8631 0.8702 HBP (ECCV18) [4] VGG16 0.8511 0.8586 DFL (CVPR18) [19] ResNet50 0.8804 0.8926 PMG (CVPR20) [20] ResNet50 0.8694 0.8825 AG Net (MIA20) [21] AlexNet+GoogleNet 0.8573 0.8644 Our Method DSOD 0.8874 0.9050 method achieves superior performance. This shows the effectiveness of the proposed approach.…”

Section: Methodsmentioning

confidence: 99%

“…In this section, we compare the performance of the proposed method with the state-of-the-art fine-grained classification methods, such as B-CNN [18], HBP [4], DFL [19], and PMG [20]. Meanwhile, the proposed method is also compared with the ensembling method [21] which is the fusion of AlexNet and GoogleNet.…”

Section: Comparison With the State-of-the-artmentioning

confidence: 99%

Learning Discriminative Representations for Fine-Grained Diabetic Retinopathy Grading

Tian¹,

Ma²,

Wen³

et al. 2020

Preprint

View full text Add to dashboard Cite

Diabetic retinopathy (DR) is one of the leading causes of blindness. However, no specific symptoms of early DR lead to a delayed diagnosis, which results in disease progression in patients. To determine the disease severity levels, ophthalmologists need to focus on the discriminative parts of the fundus images. In recent years, deep learning has achieved great success in medical image analysis. However, most works directly employ algorithms based on convolutional neural networks (CNNs), which ignore the fact that the difference among classes is subtle and gradual. Hence, we consider automatic image grading of DR as a fine-grained classification task, and construct a bilinear model to identify the pathologically discriminative areas. In order to leverage the ordinal information among classes, we use an ordinal regression method to obtain the soft labels. In addition, other than only using a categorical loss to train our network, we also introduce the metric loss to learn a more discriminative feature space. Experimental results demonstrate the superior performance of the proposed method on two public IDRiD and DeepDR datasets.

show abstract

“…For example, ImageNet is used in [8] for joint learning (Flowers: 97.7%), and pretrained sub-networks or even higher image resolution (e.g. 448×448 in PMG [17], 480×480 in GPipe [29], MC [8], etc.) is considered in improving FGVC accuracy.…”

Section: Fine-grained Image Classificationmentioning

confidence: 99%

An attention-driven hierarchical multi-scale representation for visual recognition

Wharton¹,

Behera²,

Bera³

2021

Preprint

View full text Add to dashboard Cite

Convolutional Neural Networks (CNNs) have revolutionized the understanding of visual content. This is mainly due to their ability to break down an image into smaller pieces, extract multi-scale localized features and compose them to construct highly expressive representations for decision making. However, the convolution operation is unable to capture long-range dependencies such as arbitrary relations between pixels since it operates on a fixed-size window. Therefore, it may not be suitable for discriminating subtle changes (e.g. fine-grained visual recognition). To this end, our proposed method captures the high-level long-range dependencies by exploring Graph Convolutional Networks (GCNs), which aggregate information by establishing relationships among multiscale hierarchical regions. These regions consist of smaller (closer look) to larger (far look), and the dependency between regions is modeled by an innovative attention-driven message propagation, guided by the graph structure to emphasize the neighborhoods of a given region. Our approach is simple yet extremely effective in solving both the finegrained and generic visual classification problems. It outperforms the state-of-the-arts with a significant margin on three and is very competitive on other two datasets.

show abstract

Fine-Grained Visual Classification via Progressive Multi-Granularity Training of Jigsaw Patches

Cited by 6 publications

References 37 publications

SelectAugment: Hierarchical Deterministic Sample Selection for Data Augmentation

SelectAugment: Hierarchical Deterministic Sample Selection for Data Augmentation

Learning Discriminative Representations for Fine-Grained Diabetic Retinopathy Grading

An attention-driven hierarchical multi-scale representation for visual recognition

Contact Info

Product

Resources

About