Competition Six

Link to Download Competition Files

Description

For this contest, you will be training a object detector based on YOLO (see below) to detect a custom object from a series of images.

The training data is images of cars on roads taken from a side-view-mounted camera on a car. There is one label to train on: "car" or nothing. Bounding boxes are also specified in the training data for locations of cars to train on.

Submission Requirements

Code must be in Python3

You may use any machine learning framework to train your custom detector.

Please submit code used for training and bounding box predictions (in the form outlined in the sample_submission.csv file provided in the data folder), to our email tjmachinelearning@gmail.com. The subject of your email should be NMLO_6_Lastname_Firstname or NMLO_6_Lastname_Firstname_PartnerLastName_PartnerFirstName if you are working with a partner.

We will score you on your latest submission at the end of the competition.

External Resources

What is YOLO