CS 194-26: Image Warping and Mosaicing

Kin Seng Chau

Shoot and Digitize Images

Pictures taken

lab_left lab_right

street1_left street1_right

left right

Recover Homographies, Warping, and Rectification

We compute homography H by obtaining 4 pairs of correspondence from two images through hand-picking the pixels in Adobe Photoshop, and solving a system of equations that correlate the 4 points. Both inverse and forward warping was implemented, the inverse warping fill in empty pixels using the nearest pixel while nothing was done using the forward warping. You will find the following images warped using the inverse warping.

lab_right lab_warped

street1 street1_warped

Mosaics and Blending

By computing how much each corner of the image has traslated after warping, we know how much we should move one image such that the two images overlap accurately. Since not much distortion and color difference is observed, simply taking the element-wise maximum of the the pixels work out pretty okay for my set of photos. The last set of pictures have artifacts in the center of the mosaic, which is probably due to the fact that I accidentally moved my camera slightly when I was trying to take the second picture, such that the two images weren't taken at the exact same point.

lab_left lab_right lab

street1_left street1_right street1

street2_left street2_right street2