This puzzle is a tricky one! Given no training set and no labels, can you detect starship from images? Explore traditional computer vision algorithms or experiment with unsupervised learning. Your task is to identify the body of the starship from images captured from the launch!
Feeling confused? Check out the starter kit to get some clarity!
💪 Getting Started
The images represent a starship launched from the Boca Chica Village, Texas. And your task will be to identify the body of the starship.
Use our Getting Started Notebook available here.
In this dataset, which is data.zip. It will contain over 10000 images. And the images represent the starship such as -
And your task will be to return the bounding boxes for the starship body in (x, y, w, h) format. Note that there should only be one bounding box per sample. Sample submission.csv -
|0||[213, 226, 94, 50]|
|1||[206, 194, 55, 155]|
Following files are available in the resources section:
- data.zip - [ 10000 samples ] Used for evaluation. Does not include labels.
- Creating a submission directory
- Save the submission.csv in the submission directory. The name of the above file should be submission.csv.
- 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 ├── assets │ └── submission.csv └── original_notebook.ipynb
- Zip the submission directory!
Make your first submission here 🚀 !!
🖊 Evaluation Criteria
During the evaluation, the Average Precision (AP) @[ IoU=0.50:0.50 | area= large | maxDets=100] will be used to test the efficiency of the model.
- 💪 Challenge Page: https://www.aicrowd.com/challenges/starship-detection
- 🗣️ Discussion Forum: https://www.aicrowd.com/challenges/starship-detection/discussion
- 🏆 Leaderboard: https://www.aicrowd.com/challenges/starship-detection/leaderboards