Image classification图像分类,根据各自在图像信息中所反映的不同特征,把不同类别的目标区分开来的图像处理方法。它利用计算机对图像进行定量分析,把图像或图像中的每个像元或区域划归为若干个类别中的某一种,以代替人的视觉判读。
Engkarat Techapanurak, Takayuki Okatani
We have presented a meta-approach to be used with any OOD detection method to domain shift detection, which has been poorly studied in the community
Cited by0BibtexViews12DOI
0
0
Debesh Jha, Anis Yazidi, Michael A. Riegler, Dag Johansen, Håvard D. Johansen, Pål Halvorsen
We propose LightLayers, a method for reducing the number of trainable parameters in deep neural networks
Cited by0BibtexViews8DOI
0
0
Jing Xu, Yu Pan, Xinglin Pan,Steven Hoi, Zhang Yi,Zenglin Xu
The remarkable success of ResNets is mainly due to the shortcut connection mechanism, which makes the training of a deeper network possible, where gradients can directly flow through building blocks and the gradient vanishing problem can be avoided in some sense
Cited by0BibtexViews8DOI
0
0
Arkabandhu Chowdhury, Mingchao Jiang, Chris Jermaine
We have shown that a learner built on top of a high-quality deep CNN can have remarkable accuracy, and that a learner built upon an entire library of CNNs can significantly outperform a few-shot learner built upon any one deep CNN
Cited by0BibtexViews2DOI
0
0
CVPR, (2019): 558-567
We further show that models trained with our tricks bring better transfer learning performance in other application domains such as object detection and semantic segmentation
Cited by219BibtexViews166DOI
0
0
IEEE Transactions on Geoscience and Remote Sensing, no. 9 (2019): 6690-6709
We briefly introduced several deep models that are often used to classify Hyperspectral image, including stacked auto-encoders, deep belief networks, convolutional neural networks, recurrent neural networks, and generative adversarial networks
Cited by0BibtexViews29
0
0
William H. Beluch,Tim Genewein,Andreas Nürnberger, Jan M. Köhler
CVPR, pp.9368-9377, (2018)
Through additional experiments we find that the difference in active learning performance can be explained by a combination of decreased model capacity and lower diversity of Monte Carlo Dropout ensembles
Cited by83BibtexViews9DOI
0
0
computer vision and pattern recognition, (2018)
We propose BC learning for image recognition in Section 3, explaining the relationship with BC learning for sounds
Cited by61BibtexViews10DOI
0
0
CVPR, (2017): 6450-6458
The second benefit comes from encoding top-down attention mechanism into bottom-up top-down feedforward convolutional structure in each Attention Module
Cited by1008BibtexViews152DOI
0
0
IEEE Transactions on Circuits and Systems for Video Technology, no. 12 (2017): 2591-2600
As illustrated in Fig. 3, Table III(a) and, our proposed cost-effective AL framework overcomes the compared method from the aspects of the recognition accuracy and user annotation amount
Cited by211BibtexViews106DOI
0
0
CVPR, (2017)
One characteristic that sets humans apart from modern learning-based computer vision algorithms is the ability to acquire knowledge about the world and use that knowledge to reason about the visual world. Humans can learn about the characteristics of objects and the relationships...
Cited by144BibtexViews28DOI
0
0
CVPR, (2017): 2027-2036
Extensive evaluations on NUS-WIDE, MS-COCO, and WIDER-Attribute datasets show that our proposed Spatial Regularization Net significantly outperforms state-ofthe-arts
Cited by132BibtexViews26DOI
0
0
CVPR, (2016): 2285-2294
We propose a unified convolutional neural networks-recurrent neural networks framework for multilabel image classification
Cited by591BibtexViews99DOI
0
0
IEEE Transactions on Image Processing, no. 12 (2015): 5017-5032
We proposed arguably the simplest unsupervised convolutional deep learning network— PCA Network
Cited by918BibtexViews128DOI
0
0
CVPR, (2015)
Image classification has advanced significantly in recent years with the availability of large-scale image sets. However, fine-grained classification remains a major challenge due to the annotation cost of large numbers of fine-grained categories. This project shows that compel...
Cited by555BibtexViews57DOI
0
0
IEEE Transactions on Pattern Analysis and Machine Intelligence, no. 7 (2015): 1425-1438
In the few-shots setting, we showed improvements with respect to the Web-Scale Annotation By Image Embedding algorithm, which learns the label embedding from labeled data but does not leverage prior information
Cited by314BibtexViews103DOI
0
0
ICCV, (2015)
This is not an easy problem, as we have found that naive approaches such as concatenating the GPS coordinates into the classifier, or leveraging nearby images as a Bayesian prior result in almost no gain in performance
Cited by58BibtexViews90DOI
0
0
international conference on learning representations, (2014)
The final image classification system submitted to ILSVRC2013 was composed of 10 neural networks made up of 5 base models and 5 high resolution models and had a test set top 5 error rate of 13.6%
Cited by292BibtexViews51DOI
0
0
Computer Vision and Pattern Recognition, pp.3646-3653, (2014)
We propose a principled approach for selecting a set of transformations, termed Image Transformation Pursuit
Cited by79BibtexViews65DOI
0
0
International Conference on Learning Representations, (2013)
We presented two visualisation techniques for deep classification ConvNets
Cited by2127BibtexViews182DOI
0
0