
Ball & Bat Tracking Hiring Challenge
Cricket Ball & BatTracking for international cricket dataset
Interns
Welcome to Attributics.AI Internship Hiring Challenge!
Dataset information: There are two folders within the dataset provided
1. test_videos - Contain videos on which your model needs to be tested
2. train - This dataset contains “Images” folder - which has images containing bat and ball. There is a CSV file “bat_ball.csv” which contains following columns “class (label)” “X axis (Top Left X-coordinate of the image )” “Y Axis (Top Left Y-coordinate of the image)” “width(width of the class (bat or ball))” , “height (height of the class (bat or ball))” , “name (Name of the image)”, “image_width”, “image_height”
Approach :
1) you have to detect and localize ball and bat In a given frame.
2)For this you can use any object detection model built-in pytorch or tensor-flow (yoloV4 or mobilenetssd)
3) Information about the training dataset provided is as follows
31. )There is a folder named “images” which has all the images containing bat and ball. These images are already annotated and the annotation file is a CSV file named “bat_ball” at the same level as the “images folder”
3.2) the columns of the CSV file are described above
4) once you train the models test them on the videos given in the “test_videos” folder.
5) outcome - detect bat and ball in each frame of test videos and draw a rectangular bounding box around bat and ball in each frame. Merge the images as videos. Zip your output video files & ipynb notebook with well-commented code and upload
*************************************
Submission Evaluation: Manual
Submission Grade Time: 24 Hrs - 36 Hrs.
The results will be declared when ALL the submissions have been graded.
*************************************
Please submit a single ZIP file containing test videos and IPYNB files (well commented code) in the My Submission Section of this challenge.
How to make a submission?
- Click on "My Submission"
- On the next page, click on "+ New Submission"
- Upload your ZIP in the next page and click on "Submit for Review"
- We need to see an approach of object detection with deep learning.
- This is not a challenge for building classification models using ML
Rules
- We are using automatic anti-cheat system. Our system flags submission which have low trust scores. Manually modifying your submission CSV (via comparison of any form etc.), using image comparison techniques (pixel matching, file size matching etc,), in any form, can lead to your disqualification without any notice.
- The submission should be in a proper format as described by "Submission Guidelines".
- Late submission will not be accepted beyond provided deadline (Indian Standard Time).
- Confidentiality should be strictly maintained.
Rewards:
- Invitation for interview for top candidates by Attributics.AI
- The participants with a successful submission will get a certificate of participation.
- The top 3 participants will receive a permanent place in Dockship's "Hall of Fame".
Challenge Started