In this coursework, we will be learning about image editing focusing on cloning,
gradient based editing, etc. The coursework is based on the following paper
“Poisson image editing”, P. Pérez, M. Gangnet, and A. Blake, Siggraph 2003.
Task 1 (25 points)
Select a grayscale image. Mark out a region using a polygon (you can use rpoly).
Remove the selected region and fill it in using Equation (2) in the paper. You are
solving for unknown intensity values inside the region R. Test the method in
smooth regions and also in regions with edges (high-frequency). Also report the
behavior as the size of the selected region increases.
Task 2 (25 + 25 points)
Now we are ready to try ‘seamless cloning’. The relevant Equations are (9) to
(11). Perform both versions (a) importing gradients and (b) mixing gradients.
Select images you like to edit and show interesting effects.
Task 3 (10 points)
Repeat task 2a for color images. You have to process R, G, B components
Task 4 (15 points)
Implement only one of the selection editing effects described in Section
4 of the paper. You can decide between: texture flattening, local illumination
changes, local colour changes or seamless tiling.
本网站支持 Alipay WeChatPay PayPal等支付方式
E-mail: firstname.lastname@example.org 微信号:vipnxx