Learning Deep Features for Discriminative Localization

深碍√TFBOYSˉ_ 2022-06-10 12:55 165阅读 0赞

[Learning Deep Features for Discriminative Localization论文笔记以及Caffe实现][Learning Deep Features for Discriminative Localization_Caffe]

[论文笔记 | Learning Deep Features for Discriminative Localization][_ Learning Deep Features for Discriminative Localization]

[论文笔记： Learning Deep Features for Discriminative Localization][Learning Deep Features for Discriminative Localization]

可以利用论文的思路对原始以224\*224或227\*227尺寸图像作为输入的AlexNet或VGGNet使用更高分辨率的图像作为输入，主要是为了利用原来在imagenet上预训练的模型进行finetuning。

https://github.com/hshota0530/caffe\_models

VGG的论文里还有提到用384x384，512x512的网络进行训练，然后再融合模型来提高精度，增加384x384的输入，类似于256x256，网络的输入参数大小为336x336，和256x256一样，随机截取作为输入，第一个卷积层把stride改为3就可以。512x512的网络由于输入图片太大，只做了两次实验，一次是卷积大小依旧7x7，stride改为4，不成功。一次是加入两个5x5，stride为2的卷积层，依旧不行，然后就没再实验。

https://github.com/metalbubble/CAM

https://github.com/jacobgil/keras-cam

https://github.com/slundqui/DeepGAP

https://github.com/markdtw/class-activation-mapping

https://github.com/jacobgil/pytorch-grad-cam

https://github.com/jacobgil/keras-grad-cam

https://github.com/taoyilee/Keras\_MedicalImgAI/blob/master/app/grad\_cam.py

https://github.com/insikk/Grad-CAM-tensorflow

https://github.com/adityac94/Grad\_CAM\_plus\_plus

https://github.com/conan7882/CNN-Visualization

[Visualizing and Understanding Convolutional Networks 阅读笔记-网络可视化NO.1][Visualizing and Understanding Convolutional Networks _-_NO.1]

[Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization][Grad-CAM_ Visual Explanations from Deep Networks via Gradient-Based Localization]

[凭什么相信你，我的CNN模型？（篇一：CAM和Grad-CAM)][CNN_CAM_Grad-CAM]

[论文笔记：WILDCAT: Weakly Supervised Learning of Deep ConvNets][WILDCAT_ Weakly Supervised Learning of Deep ConvNets]

Weakly supervised Localization using deep feature maps

[论文笔记： Weakly Supervised Semantic Segmentation Using Superpixel Pooling Network][Weakly Supervised Semantic Segmentation Using Superpixel Pooling Network]

[Learning Deep Features for Discriminative Localization_Caffe]: http://blog.csdn.net/leo_812/article/details/51999434
[_ Learning Deep Features for Discriminative Localization]: http://blog.csdn.net/bea_tree/article/details/51824217
[Learning Deep Features for Discriminative Localization]: https://zhuanlan.zhihu.com/p/27587399
[Visualizing and Understanding Convolutional Networks _-_NO.1]: http://blog.csdn.net/gm_margin/article/details/79335140
[Grad-CAM_ Visual Explanations from Deep Networks via Gradient-Based Localization]: http://blog.csdn.net/gm_margin/article/details/79339761
[CNN_CAM_Grad-CAM]: http://bindog.github.io/blog/2018/02/10/model-explanation/
[WILDCAT_ Weakly Supervised Learning of Deep ConvNets]: https://zhuanlan.zhihu.com/p/27766274
[Weakly Supervised Semantic Segmentation Using Superpixel Pooling Network]: https://zhuanlan.zhihu.com/p/27911739