P-CNN: Part-Based Convolutional Neural Networks for Fine-Grained Visual Categorization

Han, Junwei; Yao, Xiwen; Cheng, Gong; Feng, Xiao; Xu, Dong

doi:10.1109/tpami.2019.2933510

Cited by 89 publications

(46 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, resting-state FC is an alternative technique that can be used for biometric identification purposes [9,36,37]. Furthermore, multidisciplinary research methods rooted in neuroscience and computer science could also be extended to study the relationship of two states [38][39][40]. In the future, these results may help us predict and understand cognitive processing in patients who cannot perform a given task.…”

Section: Discussionmentioning

confidence: 99%

“…The similarity among subjects dropped from an average of 0.872 to 0.003. Furthermore, we subdivided the data into five different frequency bands: delta (0.5-4 Hz), theta (4-8 Hz), alpha (8-13 Hz), beta (13-30 Hz) and gamma (30)(31)(32)(33)(34)(35)(36)(37)(38)(39)(40)(41)(42)(43)(44)(45). The same prediction processes were separately performed for each frequency band.…”

Section: G Task-specific and Individual Differences In Activation Pamentioning

confidence: 99%

See 1 more Smart Citation

Prediction of Cognitive Task Activations via Resting-State Functional Connectivity Networks: An EEG Study

Wang

Zhang

Liu

et al. 2022

IEEE Trans. Cogn. Dev. Syst.

View full text Add to dashboard Cite

The resting state is an internal state that is closely related to neural activation and the performance of tasks. Studying the relationship between the resting and task states is helpful for understanding the organization of information processing. It remains unclear how information is translated between these two states. Methods: In this study, we focused on electroencephalography (EEG) data because its high time resolution allowed us to study processing both overall and in detail. Resting-state functional connectivity (FC) networks were constructed in the time and frequency domains. Results: FC constructed by synchronization of signals in the time domain was suitable for predicting event-related potential activation. In addition, FC measured by phase distributions had superior prediction accuracy for predicting spectral power. Conclusion: Our findings suggest that there is intrinsic organization across the two states. Furthermore, the activity flow modelled in different domains could reflect different levels of neuronal activation. Significance: Changes in neural activity across resting and task states on a sub-second time scale can be detected by EEG, which is helpful for understanding the underlying mechanisms of illness and therapeutic outcomes.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: G Task-specific and Individual Differences In Activation Pamentioning

confidence: 99%

Prediction of Cognitive Task Activations via Resting-State Functional Connectivity Networks: An EEG Study

Wang

Zhang

Liu

et al. 2022

IEEE Trans. Cogn. Dev. Syst.

View full text Add to dashboard Cite

show abstract

“…Several deep models [44]- [47] utilize visual attention to improve different remote sensing applications. Squeeze and excitation block [44], [45] improves representational capacity by utilizing channel attention for object detection. The generative adversarial network [48] is embedded with spatial attention mechanism into FCN-8s to achieve remarkable segmentation performance.…”

Section: Diverse Capsules Network Combining Multi-convolutional Layermentioning

confidence: 99%

“…In addition, to improve classification accuracy, attention mechanism has been incorporated into the deep CNN by researchers [44], [45], [62]- [64], to create more powerful discriminative representations. Visual attention mechanisms usually stimulate human visual focus on the essential objects in the scene, also known as salient objects, which are helpful for complicated scene classification [62]- [64].…”

Section: Related Workmentioning

confidence: 99%

Diverse Capsules Network Combining Multiconvolutional Layers for Remote Sensing Image Scene Classification

Raza

Sirajuddin

et al. 2020

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Remote sensing image scene classification has drawn significant attention for its potential applications in the economy and livelihoods. Unlike the traditional handcrafted features, the convolutional neural networks (CNNs) provides an excellent avenue in obtaining powerful discriminative features. Although tremendous efforts have been made so far in this domain, there are still many open challenges in scene classification due to the scene complexity with higher within-class diversity and between-class similarity. To solve the above-mentioned problems, D-CapsNet is proposed to learn the richer and more robust features for scene classification. It is an end to end network with four types of layers and incorporates visual attention mechanisms. Its diverse capsules encode different properties of complex image scenes, including deep high-level features, spatial attention based on the fusion of multilayers features, both spatial and channel attention based on high-level features, and their fusion. Experiments on three image scene datasets demonstrate that D-CapsNet outperforms other baselines and state-of-the-art methods with a significant improvement in both classification accuracy and speed.

show abstract

“…quickly gained a modest degree of application in ODRSIs. Recently, significant progress has been made on deep learning techniques [20][21][22][23], meaning the deep learning based methods now significantly outperform handcrafted features and shallow learning based methods. At present, the most prevalent deep learning based methods are anchor-based methods, which can be roughly classified into one-stage and two-stage methods according to their strategy for generating proposals.…”

Section: Introductionmentioning

confidence: 99%

Object Detection in Remote Sensing Images Based on Improved Bounding Box Regression and Multi-Level Features Fusion

et al. 2020

Self Cite

View full text Add to dashboard Cite

The objective of detection in remote sensing images is to determine the location and category of all targets in these images. The anchor based methods are the most prevalent deep learning based methods, and still have some problems that need to be addressed. First, the existing metric (i.e., intersection over union (IoU)) could not measure the distance between two bounding boxes when they are nonoverlapping. Second, the exsiting bounding box regression loss could not directly optimize the metric in the training process. Third, the existing methods which adopt a hierarchical deep network only choose a single level feature layer for the feature extraction of region proposals, meaning they do not take full use of the advantage of multi-level features. To resolve the above problems, a novel object detection method for remote sensing images based on improved bounding box regression and multi-level features fusion is proposed in this paper. First, a new metric named generalized IoU is applied, which can quantify the distance between two bounding boxes, regardless of whether they are overlapping or not. Second, a novel bounding box regression loss is proposed, which can not only optimize the new metric (i.e., generalized IoU) directly but also overcome the problem that existing bounding box regression loss based on the new metric cannot adaptively change the gradient based on the metric value. Finally, a multi-level features fusion module is proposed and incorporated into the existing hierarchical deep network, which can make full use of the multi-level features for each region proposal. The quantitative comparisons between the proposed method and baseline method on the large scale dataset DIOR demonstrate that incorporating the proposed bounding box regression loss, multi-level features fusion module, and a combination of both into the baseline method can obtain an absolute gain of 0.7%, 1.4%, and 2.2% or so in terms of mAP, respectively. Comparing this with the state-of-the-art methods demonstrates that the proposed method has achieved a state-of-the-art performance. The curves of average precision with different thresholds show that the advantage of the proposed method is more evident when the threshold of generalized IoU (or IoU) is relatively high, which means that the proposed method can improve the precision of object localization. Similar conclusions can be obtained on a NWPU VHR-10 dataset.

show abstract

P-CNN: Part-Based Convolutional Neural Networks for Fine-Grained Visual Categorization

Cited by 89 publications

References 53 publications

Prediction of Cognitive Task Activations via Resting-State Functional Connectivity Networks: An EEG Study

Prediction of Cognitive Task Activations via Resting-State Functional Connectivity Networks: An EEG Study

Diverse Capsules Network Combining Multiconvolutional Layers for Remote Sensing Image Scene Classification

Object Detection in Remote Sensing Images Based on Improved Bounding Box Regression and Multi-Level Features Fusion

Contact Info

Product

Resources

About