Uploaded March 2026 | Updated September 2026, 1 week ago

---

A new feature in Photoshop Beta (v27.5) made me try something interesting.

The new Rotate Object AI tool can generate multiple views of an object from a single image — essentially letting you rotate a 2D photo as if it were 3D.

So I ran a small experiment.
What if those AI-generated angles could be used for photogrammetry?

Here’s the workflow I tested:

1️⃣ Start with a single photo
2️⃣ Use @Adobe @Photoshop AI to generate new angles
3️⃣ Export the images
4️⃣ Import them into @RealityScanOfficial
5️⃣ Reconstruct the 3D mesh

The result?

A 3D mesh generated from images that originated from just one photo.
Of course, the more real photos you have, the better the reconstruction will be — but it’s fascinating to see how AI can help expand the dataset.

It’s still experimental, but it highlights an interesting intersection between generative AI and photogrammetry.
This could open new possibilities for rapid asset creation, concept art, and 3D reconstruction workflows.
Curious to see where this goes.


My Dell workstation: creatorfolio.co/anderson-rohr


#DellProPrecision #DellTech #NVIDIA #AI #Photoshop #Photogrammetry #RealityScan #3D #VFX #CreativeTech