search

xxlong0 / SparseNeuS

SparseNeuS: Fast Generalizable Neural Surface Reconstruction from Sparse views
MIT License
325 stars 16 forks source link

Testing on the DTU_TEST and small cavities on meshes for some scans #35

Closed erenovic closed 1 year ago

erenovic commented 1 year ago

Hello, thanks for the great work! I am currently trying to test the model after training both lod0 and lod1 as described in the repository. But for testing, the dataset appears to be similar to the sample finetuning dataset rather than the training dataset. So I decided to use the Dtu_Fit dataset class to load the DTU_TEST dataset you provided in one of the issues while using the validation code you provided in trainer_generic.py.

However, when I try to reconstruct some of the scans, I observed some small cavities as in the following screenshot:

Screenshot 2023-04-18 at 19 49 25

Can you provide some guidance about testing of the trained network on test dataset?

erenovic commented 1 year ago

Hello, the issue seems to be related to the fact that my retrained model could not achieve the same performance as the weights you provided. Everything else works fine