New Version of Dataset for IMGCOL and OBJDE

🛠 Contribute: Found a typo? Or any other change in the description that you would like to see? Please consider sending us a pull request in the public repo of the challenge here.

🕵️ Introduction

We have all seen little kids point to objects and call them out. By doing so, they are able to understand the world around them better. Even teaching kids to do this is somewhat akin to training a neural network! With this object detection challenge, can you teach an algorithm to identify objects and label them in different categories? Use our starter-code kit and teach an AI model to identify and label objects it sees?

Understand with code! Here is getting started code for you.😄

💾 Dataset

The dataset consists of images of the 5 classes Person, Clothing, Car, Plant and Footwear. The XMin, XMax, YMin and YMax coordinates of the bounding box for each of the images are available in the corresponding csv which are the normalized between 0 and 1.

📁 Files

Following files are available in the resources section:

Note: The v2 files for train and validation set are just a subset of the version 1. There is no change in the test images.

• train-v2.csv - (20000 samples) This CSV file contains labels and bounding boxes of training images ( there can be images with multiple objects! ). XMin, XMax, YMin and YMax are the coordinates of the bounding boxes normalized between 0 and 1.

• train-v2.zip - (20000 samples) The zip file containing training images.

• val-v2.csv - (2000 samples ) This CSV file contains labels and bounding boxes of validation images ( there can be images with multiple objects! ). XMin, XMax, YMin and YMax are the coordinates of the bounding boxes normalized between 0 and 1.

• val-v2.zip - (2000 samples) The zip file containing validation images.

• test-v2.zip - (10000 samples) The zip file containing testing images

• train.csv - (40000 samples) This CSV file contains labels and bounding boxes of training images ( there can be images with multiple objects! ). XMin, XMax, YMin and YMax are the coordinates of the bounding boxes normalized between 0 and 1.

• train.zip - (40000 samples) The zip file containing training images.

• val.csv - (4000 samples ) This CSV file contains labels and bounding boxes of validation images ( there can be images with multiple objects! ). XMin, XMax, YMin and YMax are the coordinates of the bounding boxes normalized between 0 and 1.

• val.zip - (4000 samples) The zip file containing validation images.

• test.zip - (10000 samples) The zip file containing testing images

• sample_submission.csv - (10000 samples) This CSV file contains a sample format of submitting the testing predictions

🚀 Submission

• Prepare a CSV containing header as [ImageID, LabelName, XMin, XMax, YMin, YMax, score] and denoting the image ids and the corresponding predicted values.

• Sample submission format available at sample_submission.csv in the resources section.

Make your first submission here 🚀 !!

🖊 Evaluation Criteria

During evaluation Average Precision (AP) @[ IoU=0.50:0.50 | area= all | maxDets=100 ] will be used to test the efficiency of the model.

🔗 Links

• 💪 Challenge Page: https://www.aicrowd.com/challenges/objde
• 🗣️ Discussion Forum: https://www.aicrowd.com/challenges/objde/discussion
• 🏆 Leaderboard: https://www.aicrowd.com/challenges/objde/leaderboards

📱 Contact

• Shubhamai