Open jingyu198 opened 2 months ago
when training, there is a bug indicating that "can not find mask file", thus I set read_mask to False, does this lead to this unreasonable result?
Fixed a severe training bug. The "zero_init_camera_projection" in 'configs/train/stage1-mix-6views-lvis.yaml' should be False. Otherwise, the domain control and pose control will be invalid in the training.
Very wonderful work and amazing demos!
I am trying to use my own data to train the model for stage 1 and use your weights as pretrained model. The only changes include pretrained model path and training data path.
However, when testing using test_mvdiffusion_seq.py, the normals seem almost the same as the RGB images. why does this happen?
![res](https://github.com/xxlong0/Wonder3D/assets/83287600/eb910c23-34ea-4566-b5f6-2b9d6823b343)