Yuting Zhang

@moh3th1 For which part of the model? If you refer to CNN, you can simply use caffe to finetune from the caffemodel that we provided. Or you can also just...

@zizhaozhang This looks like a problem in compiling the lbfgs-b module of GPML toolbox. You probably can find some useful information by referring to the README file of the GPML...

@zizhaozhang Oops ... There is a wrong link in get_voc_2007.sh I corrected in the latest commit. Thanks for pointing it out!

For Human3.6M, we also used optical flow as self-supervision (see Appendix C). The result may not be as good as those in the paper if the optical flow is not...

Sorry for the delayed response due to my recent job transition. The method is not robust to background variations for human body images (though it works for faces). The human...

Just added a Google Drive link in the readmo: https://github.com/YutingZhang/lmdis-rep/ Thanks! On Wed, Jul 20, 2022 at 10:05 PM jojolee123 ***@***.***> wrote: > Thank you for you nice work! >...

@sliftist I also often notice similar artifacts at the edges produced by the vit-h model. When using non-natural images (like 3D anime snapshots), these artifacts can sometimes become quite messy,...

Redirected from #3097 . One simple fix is to remove the CHECK statement for num(), but this will ignore the dimension mismatch error. It is probably good to generalize all...