search

VITA-Group / ABD-Net

[ICCV 2019] "ABD-Net: Attentive but Diverse Person Re-Identification" https://arxiv.org/abs/1908.01114
MIT License
306 stars 55 forks source link

How did you visualize the attention map in two branches model？ #29

Closed voyagerw closed 4 years ago

voyagerw commented 4 years ago

Hi，could you tell how did you visualize the attention map in two branches model with Grad-CAM？As far as I know，Grad-CAM script in repo“pytorch-cnn-visualizations” can just visualize the model with only one branch.

Dtappledoghuati commented 4 years ago

Hello, I would like to ask how many gpu-devices you use. I use a laptop to run the code, but I cannot set it to 0. I would like to ask your situation. Thanks.

voyagerw commented 4 years ago

Sorry， I have never trained the model， I just use it for testing performance in my pc with GTX1060(6GB)

Hello, I would like to ask how many gpu-devices you use. I use a laptop to run the code, but I cannot set it to 0. I would like to ask your situation. Thanks.

Tianlong-Chen commented 4 years ago

Hi, I will refer this method which is also adopted in our paper.

RAM visualization method from paper: Towards rich feature discovery with class activation maps augmentation for person re-identification.

This method is much simpler and effective.