fudan-zvg/PolarFormer

Issues

• How to get the cartesian output?

  #14 opened a year ago by yangH8
  1

• How to get the pre-trained weights of the backbone network？

  #15 opened a year ago by vfan26
  3

• same question about Cross-plane encoder

  #13 opened a year ago by yangH8
  0

• How much the loss can be convergence？

  #10 opened 2 years ago by mayechi
  1

• Train on 3090

  #9 opened 2 years ago by iris0329
  4

• Where is the fcos3d.pth ?

  #12 opened 2 years ago by sinsin1998
  1

• Welcome update to OpenMMLab 2.0

  #11 opened 2 years ago by vansin
  0

• Initialization weight of PolarFormer-Vov.

  #8 opened 2 years ago by Daniel-xsy
  6

• Mismatch between PolarFormer-Vov checkpoint and config file.

  #7 opened 2 years ago by Daniel-xsy
  2

• conversion of Polar coordinates and the Polar alignment module

  #6 opened 2 years ago by lwq54013
  2

• Why img_neck has no encoder layer?

  #5 opened 2 years ago by Aswhalefall
  2

• How long does it take to train the model for segmentation?

  #4 opened 2 years ago by Yiir-1
  2

• Consistency of code and paper

  #3 opened 2 years ago by yangxh11
  0

• when code will be relased

  #1 opened 2 years ago by gsqnewsix
  0

• when code will be released

  #2 opened 2 years ago by gsqnewsix
  1