Knowledge Transfer Based Fine-Grained Visual Classification

Zhang, Siqing; Du, Ruoyi; Chang, Dongliang; Ma, Zhanyu; Guo, Jun

doi:10.1109/icme51207.2021.9428148

Cited by 4 publications

(4 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A lower similarity 𝑀 represents a higher complementarity between these two features. Next, we can use the similarity 𝑀 to obtain the complementarity factor 𝐴 ∈ [0,1], 𝐴 = 𝑆𝑜𝑓𝑡𝑀𝑎𝑥(−𝑀) (13) where 𝐴 ∈ 𝑅 × represents the degree of complementarity of feature 𝑋 to feature 𝑋 . Then using the complementarity factor we can get the complementary information 𝑌 .…”

Section: 𝑀 = 𝑓(𝑋 𝑋 ) 𝑓 = 𝑋 𝑌mentioning

confidence: 99%

“…In addition to this, these three types of methods have the problem of over-enhancing some features, which leads to over-fitting of the system and poor generalization ability. In order to overcome the problem of poor generalization ability caused by over-enhancing some of the features, this paper proposes multi-feature enhancement module (MFEM) inspired by the use of erasure mechanism to force the network to focus on the potential information of the classification network [12], [13]. MFEM uses the attention mechanism to obtain an enhanced feature map at each stage of the network, and then weakens the most salient features at the next stage before obtaining an enhanced feature map.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fine-Grained Butterfly Classification Based on Multi-Feature Enhancement

Jin,

Sha,

Xie

2024

IEEE Access

View full text Add to dashboard Cite

Due to the challenges caused by the complexity and repetitiveness of discriminative features in fine-grained butterfly images, this paper proposes a fine-grained butterfly classification model based on multi-feature enhancement and multi-scale fusion to improve the accuracy of fine-grained butterfly classification. We use ResNet50 as the base network to extract features at different scales of the image. Next, the discriminative features are enhanced using spatial attention. The enhanced discriminative features are fed into the next stage of the base network, and the above operation is repeated to obtain the enhanced discriminative features at different scales. We then use the channel covariance attention module to eliminate the extra noise introduced during multi-feature enhancement. Finally, the final feature map is obtained by fusing the complementary information of different scale feature maps using the multi-scale fusion module. Ultimately, the method used in this paper achieved 96.896% accuracy on a dataset of 10 classes of finegrained butterfly classification, and the comparison experiments proved that the method used in this paper outperforms the mainstream fine-grained visual classification methods.INDEX TERMS Butterfly classification, fine grained visual classification, multiscale fusion, multi feature enhancement.

show abstract

Section: 𝑀 = 𝑓(𝑋 𝑋 ) 𝑓 = 𝑋 𝑌mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Fine-Grained Butterfly Classification Based on Multi-Feature Enhancement

Jin,

Sha,

Xie

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The WSLL can locate lesions in a weakly-supervised manner for further lesion decomposition with the CMZ strategy. C. Contrastive Multi-Zoom strategy 1) Motivation of the CMZ strategy: Extracting disentangled features can enhance recognition as not all discriminative regions (e.g., lesions) appear [18], [42], [43]. Intuitively, the various lesion scope induces significant intra-class variance.…”

Section: B Weakly-supervised Lesion Localization Modulementioning

confidence: 99%

“…More specifically, low-overlapping and disentangled inputs are more suitable for partial lesion recognition than the entangled ones. Nonetheless, the widely-used attention-guided Center Zoom mechanism only zooms in on the whole suspicious area, which increases the difficulties of feature decomposition and the probability of ignoring subtle regions [18], [43]. We delicately designed our CMZ strategy for reducing the intersection of the local-branch inputs inspired by Contrastive Crop [44], which is proposed first in self-supervised learning (SSL) for data augmentation.…”

Section: B Weakly-supervised Lesion Localization Modulementioning

confidence: 99%

CDNet: Contrastive Disentangled Network for Fine-Grained Image Categorization of Ocular B-Scan Ultrasound

Dan¹,

Li²,

Wang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Precise and rapid categorization of images in the Bscan ultrasound modality is vital for diagnosing ocular diseases. Nevertheless, distinguishing various diseases in ultrasound still challenges experienced ophthalmologists. Thus a novel contrastive disentangled network (CDNet) is developed in this work, aiming to tackle the fine-grained image categorization (FGIC) challenges of ocular abnormalities in ultrasound images, including intraocular tumor (IOT), retinal detachment (RD), posterior scleral staphyloma (PSS), and vitreous hemorrhage (VH). Three essential components of CDNet are the weaklysupervised lesion localization module (WSLL), contrastive multizoom (CMZ) strategy, and hyperspherical contrastive disentangled loss (HCD-Loss), respectively. These components facilitate feature disentanglement for fine-grained recognition in both the input and output aspects. The proposed CDNet is validated on our ZJU Ocular Ultrasound Dataset (ZJUOUSD), consisting of 5213 samples. Furthermore, the generalization ability of CDNet is validated on two public and widely-used chest X-ray FGIC benchmarks. Quantitative and qualitative results demonstrate the efficacy of our proposed CDNet, which achieves state-ofthe-art performance in the FGIC task. Code is available at: https://github.com/ZeroOneGame/CDNet-for-OUS-FGIC.

show abstract

CDNet: Contrastive Disentangled Network for Fine-Grained Image Categorization of Ocular B-Scan Ultrasound

Dan

Wang

et al. 2023

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

Knowledge Transfer Based Fine-Grained Visual Classification

Cited by 4 publications

References 19 publications

Fine-Grained Butterfly Classification Based on Multi-Feature Enhancement

Fine-Grained Butterfly Classification Based on Multi-Feature Enhancement

CDNet: Contrastive Disentangled Network for Fine-Grained Image Categorization of Ocular B-Scan Ultrasound

CDNet: Contrastive Disentangled Network for Fine-Grained Image Categorization of Ocular B-Scan Ultrasound

Contact Info

Product

Resources

About