Sangmin Woo comments

Repositories
Issues
Comments

Results 2 comments of


                                            Sangmin Woo

depth_refinement.py generates and saves completly black output image and wmf image. Can you please look into the problem.

sorry for late response. I will look into the problem asap.

Reproduce results of LVTR-CLIP

Please refer to the following configurations. ``` aux_loss=True backbone=clip bs=16 data_type=features dec_layers=4 dim_feedforward=1024 dropout=0.1 early_stop_patienc=10 enc_layers=4 eos_coef=0.1 eval_bs=16 eval_untrainedFalse hidden_dim256 input_dropout0.5 lr=0.0001 lr_drop_step=20 method=joint n_input_proj=2 nheads=8 norm_tfeat=True norm_vfeat=True num_input_frames=64 num_input_sentences=4...