CS194 Project 1: Colorizing the Prokudin-Gorskii photo collection

Background and Images

First, I implemented a single-scale version, using the SSD metric to find the best displacement of pairs of color channels (Green and Blue, Red and Blue), holding Blue fixed. Then I align Red channel and Green channel to the Blue channel. Then, to optimize this search method, I implemented pyramid speedup, continuously half the size of picture up to four times. I then used a recursion to find displacements of smaller images, align larger images by the displacement.

Simply doing this won't always yield the best alignment, so I implemented fixed-sized cropping before aligning and after stacking. The fixed size I chose after experimentation is 10 percent of image width/length.

All displacements written in format: [x, y].

Displacement G & B: [2, 5]; R & B: [3, 12]
Displacement G & B: [2, -3]; R & B: [2, 4]
Displacement G & B: [3, 3]; R & B: [3, 6]
Displacement G & B: [14, 53]; R & B: [11, 112]
Displacement G & B: [24, 49]; R & B: [56, 104]
Displacement G & B: [17, 60]; R & B: [14, 124]
Displacement G & B: [17, 41]; R & B: [23, 90]
Displacement G & B: [9, 52]; R & B: [12, 112]
Displacement G & B: [10, 82]; R & B: [13, 178]
Displacement G & B: [26, 52]; R & B: [36, 108]
Displacement G & B: [29, 79]; R & B: [37, 176]
Displacement G & B: [-2, -1]; R & B: [-6, -1]
Displacement G & B: [-3, 3]; R & B: [1, 10]