Community Explainer Prizes π | Find Teammates π―ββοΈ
Easy-2-Follow Notebooks π» | Discord AI Community π§
Introduction
High-altitude Unmanned Aerial Vehicle is revolutionary technology! The images from high-altitude UAVs can help detect fires, humans, and wildlife like mountain terrains and thick forests. The images also help in improving understanding of hard-to-reach geographical areas.
One of the biggest challenges in using high-altitude UAV images in the presence of clouds! One cannot always wait for clear skies to collect data and these clouds act as a major obstacle in getting clear images.
Your final task for Blitz X is to remove clouds from high-altitude UAV images. Check out the starter kit to make your first submission!
πͺ Getting Started
So in this challenge, the task is to remove the clouds from a high-altitude UAV Imagery.
Use our Getting Started Notebook available here.
πΎ Dataset
In this dataset, for a given data file, such as train.zip. There will be two groups of videos, clear and cloud. The number after the video name will represent the video id. Sample data format -
train
βββ clear_0.mp4
βββ cloud_0.mp4
βββ clear_1.mp4
βββ cloud_1.mp4
βββ ...
Sample Videos -
Note -
- Make sure the videos are in .mp4 format
- Make sure that there is a total of 24 frames in the video.
π Files
Following files are available in the resources section:
- train.zip - [ 2000 samples ] Used for Training. Include labels.
- partial_train.zip - [ 500 samples ] Subset of train.zip. Use for getting started!
- test.zip - [ 500 samples ] Used for evaluation. Does not include labels.
π Submission
- Creating a submission directory
- Create a clear folder inside the submission directory.
- Inside a submission directory, put the .ipynb notebook from which you trained the model and made inference and save it as original_notebook.ipynb.
- Overall, this is what your submission directory should look like
submission
βββ clear
β βββ clear_0.mp4
β βββ clear_1.mp4
β βββ ...
β βββ clear_499.mp4
βββ original_notebook.ipynb
- Zip the submission directory!
Make your first submission here π !!
π Evaluation Criteria
During the evaluation, Mean Squared Error will be used to test the efficiency of the model.
π Links
- πͺ Challenge Page: https://www.aicrowd.com/challenges/clouds-removal
- π£οΈ Discussion Forum: https://www.aicrowd.com/challenges/clouds-removal/discussion
- π Leaderboard: https://www.aicrowd.com/challenges/clouds-removal/leaderboards
π± Contact
Notebooks
3
|
0
|
|
0
|
0
|
|
0
|
0
|
|
1
|
0
|