AI HDR merge vs in-camera HDR.

Both approaches stack multiple exposures into a single image with detail in the highlights and shadows. The difference is when and where the math happens, and what the photographer can do with the result.

What in-camera HDR is

Most modern bodies expose an HDR mode. You pick the bracket spread, the camera fires three to five frames in quick succession, and the body merges them into one JPEG before writing to the card. The merge is a fixed pipeline: the camera maker decided in advance how to balance highlights, shadows, and tone curves, and that decision is baked into the file you take home.

It is fast. It is consistent within one camera body. It is also opinionated, and you cannot un-bake the result.

What we mean by AI HDR merge

We do not merge inside the camera. You shoot bracketed RAW frames as you normally would, upload them to us, and our pipeline aligns them, picks the source data per region, and produces a JPEG sized for MLS.

The merge runs on RAW data, not the camera's pre-processed JPEG. That means more headroom: the highlights and shadows we are reaching into still have all the bits the sensor captured. Nothing has been clipped or compressed yet by the camera's HDR profile.

Why we merge after the shoot

A few reasons.

• Source files are recoverable. If the merge needs adjusting later, we still have the brackets. In-camera HDR throws away that option.
• Tone profile is consistent across cameras. A property shoot might mix a Sony Alpha and a Canon R5. In-camera HDR profiles differ. Our merge produces the same tone curve regardless of which body the brackets came from.
• Window pulls are cleaner. Indoor real estate frames with bright windows are the hard case. Pulling window detail without crushing the room interior takes more headroom than a baked JPEG provides.
• Sky replacement integrates. When sky replacement is needed, it works better against a merge that still has dynamic range to match against. A flat baked JPEG limits what the replacement layer can blend into.

What this means in practice

You shoot the way you already shoot. Aperture-priority, three or five brackets at one or two stops apart, autofocus locked, tripod when you can. Don't enable the camera's HDR mode. Just shoot the brackets and let them write to the card as RAW.

Upload the brackets as a set. We figure out which frames belong together, align them, and merge.

If a shoot looks off after editing, send it back. Because we kept the brackets, we can re-merge with different priorities, not redo the shoot.

What it doesn't fix

This is not magic. AI HDR merge does not save:

• A bracket set with significant motion between frames (people walking, fans spinning fast). Alignment can only do so much.
• Frames shot at radically different white balances by accident.
• A shoot where the brackets themselves were under or overexposed across the whole set. Merge can stretch range, not invent it.

The pipeline assumes you gave it a usable bracket set. Skill on the shoot still matters more than what happens after.

Bottom line

In-camera HDR trades flexibility for speed. AI HDR merge trades a small amount of round-trip time for headroom and a consistent tone profile across every shoot. For real estate, where consistency across listings is the actual product, that is usually the trade we want.