Blurry results with mipnerf360 dataset

[Rickyeeeeee]

Thanks for the great work! But after training with mipnerf360 bicycle dataset, with command: python train.py --source_path dataset/mipnerf360/bicycle --model_path output/bicycle2 --eval --n_views 24 --depth_pseudo_weight 0.03, I got this blurry training view rendered result:

[huangyukun26]

I got the same result too, Have you tried the llff dataset?

[Deming77]

Hi, I also obtain similar results. Have you fixed this? Thank you in advance!

[Rickyeeeeee]

@Deming77 @huangyukun26 Someone told me that it should be caused by too few point clouds initialized by fuse.ply of the bicycle provided by the author. For example, the initial point cloud of the garden scene has 3 million, but the bicycle only has 200,000. Except for bicycles, there is no such problem in other scenarios.

[huangyukun26]

@Rickyeeeeee Thanks a lot！