SOLANA MEME PROJECT

GHOSTFRAME

IT WAS MOVING THE WHOLE TIME.

CA · See Token Info for the current address.

The world is full of motion our eyes never notice.

Metal bends. Leaves tremble. Surfaces respond to forces too subtle for us to see.

GHOSTFRAME is an experimental Solana meme project exploring how computer vision can reveal those hidden changes.

We take ordinary footage, isolate subtle motion, and make the results visible — so anyone can look closer at the world around them.

THE GHOST IS THE MOTION YOU MISSED.

Interactive footage and tools require JavaScript.

02 / THE REVEAL

YOU'RE NOT SEEING EVERYTHING.

A leaf can appear almost motionless while its edge changes position by tiny amounts.

Video motion magnification makes some of those changes easier to observe.

Compare the original footage with its processed version, then inspect the evidence behind the result.

TRY THE LEAF IN HUNT

03 / THE IDEA

THE WORLD HAS BEEN MOVING WITHOUT US NOTICING.

What if the things we think are completely still are moving right in front of us?

We usually trust our eyes to tell us when something moves. But our perception has limits. A camera can record subtle changes across consecutive frames that are difficult to notice during normal playback.

Those changes are what GHOSTFRAME is interested in. Not artificial movement. Not supernatural activity. Just the small physical changes that can be extracted, amplified, and examined through video processing.

The interesting part isn't making footage look dramatic.

It's discovering something that was already happening.

04 / UNDER THE FRAME

NOT MAGIC. MOTION PROCESSING.

GHOSTFRAME explores video motion magnification — techniques that use changes across video frames to reveal subtle motion.

The process depends on the quality of the footage, the selected frequency range, and the signal available for analysis.

CAPTURE
Every video is a sequence of frames. Small movements may be present even when they are difficult to perceive at normal speed.
ISOLATE
Spatial and temporal processing helps separate selected patterns of change from the rest of the footage.
AMPLIFY
The selected changes are enhanced and reconstructed into a new video, making some subtle motion easier to observe.
VERIFY
The processed result is compared with the original. Motion estimates, frequency responses, and quality checks help reveal what the analysis actually supports.

OPEN THE LAB

VIEW THE METHOD

Source frames: The leaf, metal strip, and water scenes are generated frame by frame in a browser worker.; worker.ts / frames.ts
Filtering and reconstruction: The engine builds fine and coarse luminance bands. It applies low-pass and high-pass filters forward and backward across frames, then scales and adds the filtered detail to the original image.; magnify.ts
Evidence and limits: Analysis estimates vertical edge displacement in the selected area of the source. An FFT checks for periodic motion. Weak motion, poor edge fit, or changes dominated by lighting can yield a warning or no reliable frequency.; analyze.ts / spectrum.ts

05 / THE MEME

THE GHOST WAS ALWAYS THERE.

The ghost in GHOSTFRAME isn't a supernatural character. It's a symbol for the movement nobody noticed. A tiny vibration. A shifting edge. A change hidden between frames. That's the idea behind the meme.

LOOK AGAIN is an invitation to question what seems obvious, reveal something unexpected, and share the discovery.

GHOSTFRAME brings that idea into a Solana meme community. The experiments provide something real to explore. The meme gives people a shared symbol for the discoveries they make.

TOKEN INFO

06 / BUILDER'S NOTE

WHY I STARTED GHOSTFRAME.

I started GHOSTFRAME because I became interested in something simple: how much movement happens around us that we never actually notice?

We look at a metal strip, a leaf, or an ordinary object and assume it's still.

But a video can contain information that isn't immediately obvious to our eyes. I wanted to experiment with computer vision techniques that make some of those subtle changes visible.

The first goal is straightforward: build a working motion magnification system, make the results inspectable, and let people try it for themselves.

I'm not trying to invent motion magnification. I'm exploring what happens when that technology becomes something anyone can interact with and share.

That's what GHOSTFRAME is about.

READ FULL STORY