World in Flickers: How a Camera Assembles an Image from Flashes ✎ express

Original: "ESAR: Event-Based Synthetic Aperture Reconstruction"
arXiv:2607.15073 · 2026-07-16 · arXiv non-exclusive · independent analysis · 1 min
From a stream of flickers, scientists assembled a detailed picture of the Earth's surface.

📄 Showing the "Simple" version — "Popular" is not ready yet. Add it to favorites to help prioritize it.

Unlike regular cameras, event-based ones only record moments when light becomes brighter or dimmer. They don't produce snapshots; instead, they send an endless stream of such flashes. Scientists turned this stream into a sharp picture of the Earth while flying over it on a satellite. Each flash is like a piece of a mosaic; together they form a still image. To remove noise, they use a mathematical trick familiar to astronomers: like when searching for planets around other stars using the eclipse method, where they also gather crumbs of light. Tests with a sensor on board the ISS showed: the method excellently reconstructs large details — clouds, islands, shores. Small ripples and noise are discarded. This promises sharper and faster cameras for robots and cars.

An event-based camera works like the eye's retina: it sends only changes to the brain, saving resources. Evolution 'invented' this trick millions of years ago.

In one microsecond, light travels only 300 meters, and an event-based camera can catch such a flash.

Assembling a whole from scattered scraps is a task worthy of Solaris's ocean, which communicated in fragments of images.

Scientists
Ludwig BoltzmannWilliam BoruckiJascha Sohl-Dickstein
Concepts
single pixel imaging image reconstruction synthetic data adaptive optics reanalysis real time imaging optical cavity event selection angular resolution synthetic images 3d reconstruction tensor logarithm composite analysis random laser linear inverse model