# Our standard mono+stereo model python ../train.py --model_name MS_640x192 \ --use_stereo --frame_ids 0 -1 1 # Our low resolution mono+stereo model python ../train.py --model_name MS_416x128 \ --use_stereo --frame_ids 0 -1 1 \ --height 128 --width 416 # Our high resolution mono+stereo model python ../train.py --model_name MS_1024x320 \ --use_stereo --frame_ids 0 -1 1 \ --height 320 --width 1024 \ --load_weights_folder ~/tmp/MS_640x192/models/weights_9 \ --num_epochs 5 --learning_rate 1e-5 # Our standard mono+stereo model w/o pretraining python ../train.py --model_name MS_640x192_no_pt \ --use_stereo --frame_ids 0 -1 1 \ --weights_init scratch \ --num_epochs 30 # Baseline mono+stereo model, i.e. ours with our contributions turned off python ../train.py --model_name MS_640x192_baseline \ --use_stereo --frame_ids 0 -1 1 \ --v1_multiscale --disable_automasking --avg_reprojection # Mono+stereo without full-res multiscale python ../train.py --model_name MS_640x192_no_full_res_ms \ --use_stereo --frame_ids 0 -1 1 \ --v1_multiscale # Mono+stereo without automasking python ../train.py --model_name MS_640x192_no_automasking \ --use_stereo --frame_ids 0 -1 1 \ --disable_automasking # Mono+stereo without min reproj python ../train.py --model_name MS_640x192_no_min_reproj \ --use_stereo --frame_ids 0 -1 1 \ --avg_reprojection