Yes—this is a real technique, generally called super-resolution or pixel-shift/multi-shot imaging.

The idea is that tiny sub-pixel shifts between frames can let software estimate detail more finely than a single capture, especially when the subject and camera are very stable. It works best with static scenes, good alignment, low noise, and multiple frames. In practice, combining several shots often improves detail and reduces noise; the resolution gain is limited and won’t magically create unlimited detail beyond what the lens, focus, motion blur, and sensor sampling allow.

There are two common forms:

• Natural multi-frame super-resolution: software aligns separate exposures taken hand-held or on a tripod.
• Pixel-shift modes: some cameras intentionally move the sensor by precise tiny amounts and combine the frames.

It tends to fail or produce artifacts when anything moves between shots—subject motion, camera shake, changing light, leaves, water, people, etc. It’s most useful for architecture, products, artwork, landscapes in calm conditions, and reproduction work.

So yes: the concept is valid, and software can do it, but results depend heavily on scene stability and capture quality.