Andréas Meuleman
Andréas Meuleman
Thank you for sharing, 1. We might be able to obtain sharper results simply by tweaking parameters. I would suggest increasing the camera translation learning rate e.g. adding `--lr_t_init 1e-3`...
I think so
Hi, 1. This is only used for visualization. Use what you believe is clearer. 2. MLPRender_Fea_late_view can avoid overfitting the view-dependent component to the input views and help extrapolation when...
Hi, Instructions to prepare the dataset are available in the [README](https://github.com/facebookresearch/localrf#preprocessing). It only requires sorted image files in `${SCENE_DIR}/images`. `sparse/` and `transforms.json` are not needed when optimizing with default parameters.
The FoV depends on your focal length and sensor size. What camera are you using?
Hi, Yes previously optimized models should be stored in cpu memory. I will double check. What gpu are you using? How many frames are being considered?
I have optimized for longer sequences with 24 GB gpus. Would you mind sharing logs? At which point does it crash?
That is odd. In GPU memory, there should only be more poses, which are tiny. Do you know if it crashes during training or testing? (it render some test frames...
OK, I'll recheck and make sure there is nothing unused in GPU memory this afternoon.
I now delete optimizers and compute the alpha mask on CPU. Please let me know if the issue remains. See https://github.com/facebookresearch/localrf/commit/3905e3988e6f0e977a625b8f1f3710e90442f06b