search

googleinterns / IBRNet

Apache License 2.0
486 stars 52 forks source link

Questions about finetuning. #16

Closed pansanity666 closed 11 months ago

pansanity666 commented 1 year ago

Hi,
I have several confusions about the finetuning for the Generalizable NeRF task (IBRNet, MVSNeRF, etc):

  1. How many frames are used during finetuning? Only the reference views or the whole test set of the new scene?
  2. If the latter, the metrics are calculated with the SEEN frames during finetuning. Is it reasonable?

  3. Should I lower down the LR during finetuning?

    Best,

qianqianwang68 commented 11 months ago

Hi, the finetuning only uses the reference views (the test views are kept held out). Yes, you could try lowering the learning rate during fine-tuning and see if that leads to better results.

pansanity666 commented 11 months ago

Hi, the finetuning only uses the reference views (the test views are kept held out). Yes, you could try lowering the learning rate during fine-tuning and see if that leads to better results.

Thanks.