albedo is a const color all the time

[Easonzero]

RT

training command:
python3 tridmm/main_non_linear_3DMM.py --batch_size 16 --sample_size 128 --learning_rate 0.001 --ouput_size 224 --gf_dim 32 --df_dim 32 --dfc_dim 320 --gfc_dim 320 --z_dim 20 --c_dim 3 --is_using_landmark True --shape_loss l2 --tex_loss l1 --is_using_recon True --is_using_frecon True --is_partbase_albedo True --is_using_symetry True --is_albedo_supervision True --is_batchwise_white_shading True --is_const_albedo True --is_const_local_albedo True --is_smoothness True --gpu 0 --checkpoint_dir ckpt/ckpt_3dmm/ --samples_dir samples/tridmm/ --is_train True

[huyanfei-cqupt]

RT

training command:
python3 tridmm/main_non_linear_3DMM.py --batch_size 16 --sample_size 128 --learning_rate 0.001 --ouput_size 224 --gf_dim 32 --df_dim 32 --dfc_dim 320 --gfc_dim 320 --z_dim 20 --c_dim 3 --is_using_landmark True --shape_loss l2 --tex_loss l1 --is_using_recon True --is_using_frecon True --is_partbase_albedo True --is_using_symetry True --is_albedo_supervision True --is_batchwise_white_shading True --is_const_albedo True --is_const_local_albedo True --is_smoothness True --gpu 0 --checkpoint_dir ckpt/ckpt_3dmm/ --samples_dir samples/tridmm/ --is_train True

hi, I want to ask that the value of ‘self.gfc_dim’ set by you? Thanks.

[tranluan]

Hi, I provide the command in the main page.
I think the problem maybe using --is_const_local_albedo True too early can be harmful. So I suggest set it to False during pretrain.