inspire-group / detectorguard Goto Github PK

Hello,I appreciate that you guys could open your paper source code. There are some questions at below puzzled me when i used your algorithms to defend adversarial patch in some datasets. The officially designed coco dataset cannot be used for image classification tasks; However, you perform image classification of the coco dataset on bagnet, get local_features, and then get the objectness map; So, How do you implement this image classification task firstly? I hope you could help me solve this question. Thank you so much! !

大佬，你的对抗样本（文中提到的500张都是怎么生成的呀），是使用的哪个攻击算法呀，是自己手动生成的嘛？

Hi im using train_bagnet script code to train for my own traffic sign dataset but the loss didn't seem to decrease much and bagnet33 keeps predicting objects as background after few epochs. I already find New feature map size of my input using bagnet33. What did i do wrong

Hello, Thanks for your open-source code.

I want to confirm that did you actually use the generated adversatial example to test the defense algorithm in your experimental part? In the codebase, I can't find the ralated code implementation of localized patch hiding attacks.

I want to use the image with the adversarial patch to test your defense, could you also provide the implementation of patch hiding attacks with different patch locations according to your paper in your codebase?