Rules

Phase III (Main Task):

  • There is no restriction on the type of algorithm/model you are using. Any algorithm, irrespective of whether it utilizes deep learning or not, is allowed.
  • Pre-trained models are allowed in the competition.
  • Since Data Contribution Phase has already been completed, the participants are restricted to train their algorithms on the provided Road Data train sets. Collecting additional data for the target attribute labels is not allowed. If you do so, please mail to the organizers. However, it is permitted to increase training images using data augmentation, GANs etc. artificially.
  • Images in this dataset are available under the Creative Commons Attribution-ShareAlike 4.0 International License (CC BY-SA 4.0).
  • We expect you to respect the spirit of the competition and do not cheat. Hand-labeling is forbidden.
  • There is no restriction on the size of the team.
  • Multiple submissions are allowed, and a team can submit up to five csv files every day for judging.
  • If the performance of the algorithms submitted by two different groups is found to be the same, the one who submitted earlier would be ranked higher.
  • The top 10 rankers would be invited to submit the paper.
  • The ranks would be based on the performance of your algorithms. However, the quality of technical reports may affect the ranking decision.
  • Even if you are submitting a paper with similar content to other workshops, if you add a 30% extension, you can submit through our link. However, we expect you to mention this point at the time of submission exclusively.
  • Please submit source code and trained model in executable form before the deadline. IPython Notebook is desirable for the source code submission, but any programming languages are acceptable. The organizers will verify the reproducibility of the algorithm and determine the final winner.


Phase I (Data Contribution):

  • CRDDC is an open challenge with minimal restrictions. Anyone having the road damage data (Natural/Artificial) may participate.
  • Participants need to register on the challenge website, and may participate individually or as a team. For team participation, the team leader should be clearly identified and he/she would be responsible for future communication related to the submission.
  • In phase I, participants need to provide complete information related to the proposed dataset using the questionnaire provided on the challenge website (please refer to the forms in Submission section). Additional report may be submitted if the participants want to include more information (highly encouraged).
  • It is not necessary to submit information of only own dataset. Information related to any other dataset available to participants (in public/private domain) may also be contributed. However, in case of data belonging to others, participants need to make sure that they have necessary rights to share the information and/or the dataset. Also, in this case, the participants will be given credits as “Data Recommender” in place of “Data Contributors”.
  • Multiple submissions are allowed and encouraged for phase I. For instance, if a team wish to register as “Data Contributor” for their dataset, they may also act as “Data Recommender” for some other dataset, that they feel would be suitable for inclusion in CRDDC.
  • Participants may use CRDDC as a platform to release their dataset (if not previously released). However, if the participants are willing to continue keeping the data private, they may still participate as “Information Provider” by providing information related to their dataset.
  • The teams will be assigned credits in the following order: "Data Contributors (contribute data and associated information)" > "Information Providers (provide information, keeping the data private)" > "Data Recommenders (recommending inclusion of data belonging to others in CRDDC by providing relevant information)".
  • In case, same data is recommended by multiple teams, the team providing more information will be ranked higher.