Keunhong Park

Results 40 comments of Keunhong Park

I just ran into this issue as well. It'd be great to have it merged.

Hi Dogyoon. 1. We don't use NDC coordinates as NeRF does for some of the LLFF scenes. We use normal Euclidean coordinates, in which case the specific coordinate convention doesn't...

Hi Chilson. How small are your outputs? Some of the datasets have smaller resolutions (the ones taken as a video rather than a collection of photos).

You're likely running out of GPU or system memory. If it hangs for 5 hours maybe you ran out of system memory and it's swapping?

Hi, sorry for the late response. This is a very good question. Our method uses observation-frame view directions which remain correct after deformation. With the appearance latent code, this provides...

Hi, I'm not sure I completely understand the question. Are you using COLMAP to do the camera registration? People tend to me, and even if you try really hard to...

Here you go! https://github.com/google/hypernerf/blob/main/notebooks/figures/nerfies_2d_experiments.ipynb

Did you square the residual before passing it to the function?

Hi, the weights of the last layer in the deformation MLP are initialized close to the identity, since you want it to stay close to the identity unless it needs...

It seems like your cameras do not match your images for some reason. They are quite small so it might be a rounding issue.