Pilars of Memory

For many of us, our memories rest quietly as the camera rolls on our phones, stored as silent streams of 1s and 0s. We scroll through them, familiar faces and moments, but beneath those images lies something deeper: raw data.

This project explores what happens when we experience those memories in their purest form. Instead of seeing the photos, I’ve transformed them into fields of light pillars, each representing a random image from my phone’s library. My goal was to see just how far I could compress these photos and still recover something recognizable—an experiment in the fragility of data.

Using discrete cosine transform (DCT) compression, I reduced each photo to a 50 x 50 matrix, compressing the original down to a grayscale, low-resolution version. The result may be a far cry from the original image, but it still holds onto something—an echo of a memory.

Each matrix was then transformed into a 3D patch of light, with every pillar’s height reflecting the values in the data. I added another layer by adjusting the light intensity of each pillar to match the grayscale tones of the image. The vagueness of these images are intentional, but still, from above, you can make out a faint shape—maybe a face, a person, or a place—emerging from the glowing landscape. It's a fragile reminder of the original photo, where even the smallest alteration can distort it beyond recognition.

There’s beauty in seeing these memories reduced to their most essential form, yet still holding onto their identity—just barely.