AI Blitz XI: Completed #educational #blitz Weight: 25.0

Welcome to AI Blitz XI! ๐Ÿš€ |  Starter Kit For This Challenge! ๐Ÿ› 

Community Contribution Prizes ๐Ÿ““  |  Find Teammates ๐Ÿ‘ฏโ€โ™€๏ธ

Easy-2-Follow Notebooks ๐Ÿ’ป   | Discord AI Community ๐ŸŽง

๐Ÿ”ฅ Introduction

Radar detectors, along with cameras, help navigate the car when visibility is poor. In tough weather conditions like fog or heavy rains, these sensors help in identifying the surrounding environment. They send pulses of radio waves to locate an object and send back signals about the speed and location of that object.

With radar data from your car, can you detect different vehicles around you? This starter kit will help beginners with object detection.

โœ”  The Task

The challenge is to use the images dataset to build an automated algorithm to detect individual pedestrians and other kinds of vehicles and predict the bounding boxes:

In machine learning terms: this is multi-class object detection.

๐Ÿš€ Getting Started

Make your first submission using starter kit. ๐Ÿš€

๐Ÿ’พ Dataset

The given dataset contains the images and the labels in MS COCO json format. There are over 4 classes, bicycle, motorcycle, passenger car, and person. 

๐Ÿ“‚ Files

Following files are available in the resources section:

  • train.zip    ( 3k samples )  -  Contains images for the training set. 

  • train.json  ( 3k samples )  - This is the train annotations in MS-COCO format.

  • test.zip  ( 1samples )  -   Contains images for the testing set. 

๐Ÿ“ฎ Submission Format

Make your first submission using the starter kit ๐Ÿš€

1. Create an empty list
2. Generate the predictions of individual image 
3. Add a dictionary in the list with the following keys & values:
   image_id : Id of the image you predicted, it should be +1 the file id. so forex. ( if the file name is 456.jpg, the image_id will be 457 )
   category_id  : Id of the predicted class, below the classes and its corresponding class.                

             bicycle : 0
             motorcycle : 1
             passenger_car : 2
             person : 3

    bbox  : Predicted bounding boxes in (x, y, w, h) format. 
    score   : Probability of the prediction
4. Save the list as a submission.json file.


Here show your JSON file should look like ( small sample ) : 

    "image_id": 1, 
    "category_id": 3, 
    "bbox": [649.07568359375, 342.45306396484375, 15.56268310546875, 35.48431396484375], 
    "score": 0.9439834356307983

    "image_id": 1, 
    "category_id": 2, 
    "bbox": [445.884765625, 303.1158752441406, 21.221038818359375, 8.73126220703125], 
    "score": 0.8646438121795654
    "image_id": 2, 
    "category_id": 0, 
    "bbox": [337.07666015625, 318.928955078125, 12.354644775390625, 17.37786865234375], 
    "score": 0.9342217445373535
    "image_id": 3, 
    "category_id": 1,
    "bbox": [594.2127685546875, 311.02813720703125, 27.33331298828125, 18.92327880859375],
    "score": 0.9407286643981934
    "image_id": 3, 
    "category_id": 2, 
    "bbox": [733.0745239257812, 299.1311340332031, 45.58917236328125, 22.581451416015625],
    "score": 0.8698933720588684


๐Ÿ“จ How to submit

  1. Save the submission.json in the assets directory. The name of the above file should be submission.json.

  2. Inside a submission directory, put the .ipynb notebook from which you trained the model and made inference and save it as notebook.ipynb.

  3. Zip the submission directory

  4. Overall, this is what your submission directory should look like -

    โ”œโ”€โ”€ assets
    โ”‚   โ””โ”€โ”€ submission.json
    โ””โ”€โ”€ original_notebook.ipynb
  5. Make your first submission here.

๐Ÿ–Š Evaluation Criteria

The evaluation metrics for this competition is Average Precision ( Primary Score )  @ IoU=0.50:0.95  area  all and maxDets set to 100.  

๐Ÿ“ฑ Contact

If you have any questions, consider posting on the Blitz 11 Community Discussion board, or join the party on our Discord!


See all
Plot BoundingBoxes with Matplotlib and OpenCV
11 days ago
[Getting Started Notebook] Object Detection
27 days ago