GANWANSHUI
GANWANSHUI
does anyone get the usable pretrained weight?the link does'n work,so sad!
> Thanks for you reply~~~ Did you tried it on DDAD or Nuscenes? I tried on Nuscenes and used the ground truth pose instead of PoseNets. In every iteration, both...
> I used your submission.py to run the KITTI2012 dataset, but when I submit it to the KITTI benchmark, the website always reminds me something wrong with the data format....
> Hi @Renatusphere : > > The depth in the `trainer.py` is normalized and does not reflect the absolute scale. However, the pose net and the depth net are sharing...
Thanks a lot for the great work! I wondering how to obtain ProjectionMatrix, directly use intrinsic from the sensor, or calculate from the FOV, thanks in advance if you can...
Thank you for the reply. Would it be possible for you to share this piece of code or details so that I can begin the study with those datasets? Thank...
Sure, thank you!
Hi, we downsample the ply files into PCB to reduce the computational cost. I have uploaded the PCB model to: https://www.dropbox.com/sh/zxq5lx71zpq4nts/AAALVgeSvszpHEy8CUBr8iala?dl=0. Thanks a lot for the reminder!
We did not retrain the model with ADD(S) loss. If you mean the A(M)GPD loss, what I could find in the training info about the proposed A(M)GPD loss (GADD) at...
Hi, thanks a lot for the nice work. Any update about the open-source code? Thank you!