Poisson Image Blending

An implementation of Poisson Image Blending for CS 70, Foundations of Applied Computer Science and Machine Learning. The project explores gradient-domain image processing, a technique with broad applications including blending, tone-mapping, and non-photorealistic rendering.

The primary goal is to seamlessly blend an object or texture from a source image into a target image. Rather than copying pixel values directly (which creates visible seams), the approach leverages the insight that human vision is more sensitive to image gradients than overall intensity. The algorithm finds values for pasted target pixels that maximally preserve the gradient of the source region without changing background pixels — so a green hat could turn red, but it still looks like a hat.