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

Did you know that until the 1890s, most photos were hand-colored, after the image was clicked? And now in 2021, we are using AI to color the photos!

Given a black and white photo, train your model to color it! Use our starter-code kit and make your first AI model that magically colors the images!📷

💾 Dataset

The dataset consists of black and white images and there corresponding colored images of size (512,512). The dataset is divided into train and validation sets. The train_black_white_images.zip contains the black and white images of the train set and train_color_images.zip contains the corresponding color images. Similarly for validation_black_white_images.zip and validation_color_images.zip contains b/w and color images for the validation set. 

Each image needs to be converted from grayscale to RGB color mode.

📁 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_color_images-v2.zip - (20000 samples) This zip file contains the color training images.

• train_black_white_images-v2.zip - (20000 samples) This zip file contains the gray images, the image filenames is same throughtout the train_color_images-v2.zip and train_black_white_images.zip.

• validation_color_images-v2.zip - (2000 samples) This zip file contains the color validation images.

• validation_black_white_images-v2.zip - (2000 samples) This zip file contains the color validation images.

• test_black_white_images-v2.zip - (5001 samples) This zip file contains the gray images for testing.

• train_color_images.zip - (40000 samples) This zip file contains the color training images.

• train_black_white_images.zip - (40000 samples) This zip file contains the gray images, the image filenames is same throughtout the train_color_images.zip and train_black_white_images.zip.

• validation_color_images.zip - (4000 samples) This zip file contains the color validation images.

• validation_black_white_images.zip - (4000 samples) This zip file contains the color validation images.

• test_black_white_images.zip - (5001 samples) This zip file contains the gray images for testing.

🚀 Submission

• Prepare a zip file containing RGB testing images and filenames corresponding with, make sure there is a total of 5001 images! Also do make the zip contains the folder test_back_white_images which contains the images in .jpg.

•     ├───submission.zip
          ├───test_color_images
                  |───848d938b566f5e9c.jpg
                      dab3e432f1ee4ae4.jpg
                      88151486e9421f19.jpg
  

   

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

Make your first submission here 🚀 !!

🖊 Evaluation Criteria

During the evaluation, the average [Mean Squared Error] will be calculated over all the testing images. np.mean((real_img - predicted_img)**2) is the code for calculating MSE for images.

🔗 Links

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

📱 Contact

• Shubhamai