Loading
Cookies help us deliver our services. By using our services, you agree to our use of cookies. Learn more

crowdAI is shutting down - please read our blog post for more information

ImageCLEF 2019 Security - Stego image discovery

Identify stego images


Completed
31
Submissions
59
Participants
1635
Views

ground_truth_training_set_task_2

Posted by ML_Guy 8 months ago

Hi,

After opening the ground truth file for the task 2, I have found out that it contains 1 0922 values separated by semicolons. Is this file the right one? Because given that we have 1000 imagges in the training set, I was expecting 1000 labels (binaries) as ground truth. Are we supposed to use an unsepervised approach given that the ground truth is not useful as it’s? Is the ground truth file containing the binaries but need to be decoded first?

Posted by Konstantinos Karampidis  7 months ago |  Quote

@ML_Guy Hello The file is the correct one. It contains 1000 filenames separated by semicolons. Have you downloaded the correct one? If you have any further questions don’t hesitate to contact us. Thank you for registering into our competition and we are expecting your submission