monosdf icon indicating copy to clipboard operation
monosdf copied to clipboard

Published 20 hours ago • verified publisher icon autonomousvision

Problems occurred while training my own dataset

Open wujinhu1999 opened this issue 2 years ago • 5 comments

Output during training： merge_410_6 image The configuration I use is dtu_mlp_fullres_allviews.conf, which is a dataset around the brain without a mask image. I did not run scannet_to_monosdf.py, I changed img_res in dtu_mlp_fullres_allviews.conf to [384, 384].

wujinhu1999 avatar Oct 21 '23 09:10 wujinhu1999

How many images do you use for training?

niujinshuchong avatar Nov 01 '23 23:11 niujinshuchong

How many images do you use for training?

hi,I use 68 images for training.

wujinhu1999 avatar Nov 02 '23 01:11 wujinhu1999

Could you first check if your camera poses are correct. Other than that, maybe you could try to use a smaller sphere at initialization.

niujinshuchong avatar Nov 02 '23 12:11 niujinshuchong

Could you first check if your camera poses are correct. Other than that, maybe you could try to use a smaller sphere at initialization. I would like to ask how to use a smaller sphere at initialization and why do that.

wujinhu1999 avatar Nov 02 '23 12:11 wujinhu1999

I also has the same problem, did you solve the problem?

ansj11 avatar Mar 20 '24 08:03 ansj11