• Any of the following Kaggle Contests:
  • https://www.kaggle.com/c/whale-categorization-playground
  • https://www.kaggle.com/c/imaterialist-challenge-fashion-2018
  • https://www.kaggle.com/c/imaterialist-challenge-furniture-2018
  • https://www.kaggle.com/c/landmark-retrieval-challenge
  • https://www.kaggle.com/c/landmark-recognition-challenge
• Implement any approach for Intrinsic Image Decomposition (possible dataset: http://www.cs.toronto.edu/~rgrosse/intrinsic/).
• Implement an approach for shadow detection: Try to find cast shadows in outdoor images. Possible Dataset: http://aqua.cs.uiuc.edu/site/projects/shadow.html
• Creating adversarial examples to fight classification (Description: https://blog.openai.com/adversarial-example-research/).
• Tracking: Try to track players and the ball in a sports video.

Report Requirements:

• Names of everyone in your group.
• Introduction (problem –Why, with real‐world applications), and one illustrative example output.
• Related Work (What has been done, with a few references)
• Your Approach (How you are going to do it, with algorithms, equations, figures)
• Your Implementation and Analysis (What you do, with images, tables, and numbers)
• Your Conclusions (Itemized conclusions, observations and discussions)