Skip to content

CrackIL

Directory actions

More options

Directory actions

More options

Latest commit

 

History

History

CrackIL

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
images
images
 
 
model
model
 
 
model_parameters
model_parameters
 
 
utils
utils
 
 
README.md
README.md
 
 
bdd100k.py
bdd100k.py
 
 
check_dataset.py
check_dataset.py
 
 
domain_shift.py
domain_shift.py
 
 
domain_shift.sh
domain_shift.sh
 
 
eval.py
eval.py
 
 
eval.sh
eval.sh
 
 
facial_age_estimation.py
facial_age_estimation.py
 
 
fashion_mnist_test_images.npy
fashion_mnist_test_images.npy
 
 
fashion_mnist_test_labels.npy
fashion_mnist_test_labels.npy
 
 
fashion_mnist_train_labels.npy
fashion_mnist_train_labels.npy
 
 
homework.out
homework.out
 
 
homework3.py
homework3.py
 
 
homework4_part2.out
homework4_part2.out
 
 
homework4_part2.py
homework4_part2.py
 
 
iouEval.py
iouEval.py
 
 
metrics.py
metrics.py
 
 
read_bank.py
read_bank.py
 
 
sam.py
sam.py
 
 
sam_bank.py
sam_bank.py
 
 
test.sh
test.sh
 
 
test_target.py
test_target.py
 
 
test_target_label.py
test_target_label.py
 
 
train.py
train.py
 
 
train.sh
train.sh
 
 
train_ce.sh
train_ce.sh
 
 
train_kld.sh
train_kld.sh
 
 
train_step2.py
train_step2.py
 
 
train_step2.sh
train_step2.sh
 
 
train_step2_source_as_target.py
train_step2_source_as_target.py
 
 
train_step2_with_target.py
train_step2_with_target.py
 
 
train_step2_without_kld.py
train_step2_without_kld.py
 
 

README.md

Crack Detection Module

We are segmenting the cracks using unsupervised domain adaptation approach.

fig1

Instruction to run the code :

  1. To segment the given images :
cd CrackIL 
python3 test_target.py
  1. To get the accuracy with the given images and label :
cd CrackIL 
python3 test_label.py

Note : To give the correct path of images folder and labels folder, check the passed arguments in above python file