GitHub

This is a pytorch implementation of the paper

We validate the proposed method with the two public dataset, SUN RGBD and NYU V2

cd dataset

Then, run main.py

Name		Name	Last commit message	Last commit date
Latest commit History 3 Commits
.vscode		.vscode
dataset		dataset
loss		loss
models		models
util		util
FT_ignite.py		FT_ignite.py
FT_lift_ignite.py		FT_lift_ignite.py
README.md		README.md
__init__.py		__init__.py
correlative_finetune.py		correlative_finetune.py
correlativeloss.py		correlativeloss.py
correlativeloss.pyc		correlativeloss.pyc
finetunemodel.py		finetunemodel.py
getFinefeature.py		getFinefeature.py
getfea.py		getfea.py
getfeature_rgb_lift.py		getfeature_rgb_lift.py
lift_finetune.py		lift_finetune.py
liftloss.py		liftloss.py
liftloss2.py		liftloss2.py
lxdliftloss.pyc		lxdliftloss.pyc
main.py		main.py
one.py		one.py
sun_class_weights.npy		sun_class_weights.npy
sun_main.py		sun_main.py
sun_test.py		sun_test.py
sun_test.pyc		sun_test.pyc
test.py		test.py
test.pyc		test.pyc
two.py		two.py