Experiments on ScanNet - Githubissues

facebookresearch / consistent_depth

We estimate dense, flicker-free, geometrically consistent depth from monocular video, for example hand-held cell phone video.

MIT License

1.61k stars 236 forks source link

Open csBob123 opened 3 years ago

csBob123 commented 3 years ago

Hi, Thank you for sharing the code. Could I know which method is used for estimating the camera pose on ScanNet?

I checked the paper, but I cannot find the details.

So sorry for my troubles.

Thank you for your attention