I want to use computer vision techniques to find a book in a random background image, figure out how it's rotated, and then overlay a new book onto it. Here are the steps I plan to take:

  1. Use object detection algorithms to locate the book in the image and determine its dimensions and position relative to the top left corner of the image.
  2. Figure out the orientation of the book so that I can rotate the new book to match it.
  3. Create a new book model that takes into account the translation and rotational angles I found in steps 1 and 2, and then overlay it onto the old book to replace it.

I'm not sure how to approach steps 2 and 3, so any advice would be appreciated. Thank you!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.