search

aimerykong / Low-Rank-Bilinear-Pooling

CVPR2017 - an ultra-compact bilinear model for fine-grained classification
https://www.cs.cmu.edu/~shuk/lr_bilinear.html
149 stars 36 forks source link

The demo4_train_LRBP does not include the file of bisvm_posneg_UUreg_forward/backward ? Can you provide? #3

Closed ZBYuan0813 closed 4 years ago

aimerykong commented 6 years ago

I've uploaded the two file to this folder --

https://github.com/aimerykong/Low-Rank-Bilinear-Pooling/tree/master/demo4_train_LRBP/froBiCls

Thanks.

On Tue, Oct 31, 2017 at 4:16 AM, zanghao2 notifications@github.com wrote:

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/aimerykong/Low-Rank-Bilinear-Pooling/issues/3, or mute the thread https://github.com/notifications/unsubscribe-auth/AGKZJLAqxPlwl55uw2lZih-XnVcIE57vks5sxwGBgaJpZM4QMk9O .

ZBYuan0813 commented 6 years ago

Thank you !

shiyuanyin commented 5 years ago

@aimerykong Hi ,thank you for your paper work ,is great. but I have some question to ask for your help.

  1. I still don't know the vgg16 param m =100,r=8, how to get the m .If I change the model or dataset, how should I calculate these values?
  2. I'm sorry your formula calculation is a little hard for me to understand. Could you provide me some easy understand imformation or paper? I want to know how to infer formulas.

For our model, we set m= 100 and r= 8, corresponding to the reduced feature dimension and the rank of our low-rank classifier, respectively

Thank you for your reply(If you are more familiar with Chinese ，You can reply in Chinese for me ）

shiyuanyin commented 5 years ago

@aimerykong I think your visilization is cool ,I also want to realise it.Could you provide me with relevant information and code? 5.5. Qualitative Visualization To better understand our model, we adopt three differ-ent approaches to visualizing the model response for spe-cific input images.