Gustav The word sounds technical. The concept is surprisingly straightforward.
Photogrammetry is the process of taking a large number of overlapping photographs of a real object or space - from every angle - and using software to reconstruct it as a precise, textured 3D model. You end up with a digital twin of whatever you pointed the camera at: a product, a room, a building, an archaeological site, a person.
That 3D model can then be used in renders, in real-time applications, in VR experiences, in digital archives, or as a base for further 3D work. It captures something a manually modelled object cannot: the reality of the original. Every scratch, every grain in the wood, every irregularity in the surface is preserved exactly as it was.
Where photogrammetry is actually being used in 2026
Product visualisation. E-commerce businesses use photogrammetry to create 3D models of existing products - especially complex or organic shapes that would be expensive to model manually. From that model they can generate renders for any angle, in any context, without further photography.
Architecture and real estate. Capturing as-built conditions of spaces for renovation planning, interior design, or marketing. Walk a camera through a finished apartment, process the footage, and deliver a navigable 3D model that gives remote buyers or designers an accurate sense of the space.
Cultural heritage and preservation. Museums, galleries, and conservation institutions use photogrammetry to create permanent digital records of physical objects - sculptures, artefacts, historical buildings. If something is damaged or destroyed, the digital twin survives. I worked on exactly this kind of project for historical preservation, and the precision you can achieve with modern equipment and software is remarkable.
Game and film production. Photogrammetry is now a standard part of the pipeline for high-end games and film visual effects. Real locations and props are scanned and brought directly into 3D environments, saving enormous amounts of manual modelling time while achieving a level of realism that handcrafted models rarely match.
Industrial documentation. Factories, equipment, and infrastructure captured for maintenance planning, insurance documentation, or digital twin applications. The ability to have a precise 3D record of a physical installation without expensive laser scanning is a significant cost reduction for many industries.
What it actually costs
The honest answer is: it depends heavily on the size and complexity of what you are capturing, and on what you need the output for.
At the low end - a single small product, captured in a controlled environment, processed into a clean model for renders - you are typically looking at a few hundred euros for a professional result. The capture itself is fast. The processing and cleanup are where the time goes.
For larger subjects - a full interior space, a vehicle, a complex industrial object - the time to capture increases significantly, and so does the processing work. A thorough scan of a medium-sized room, processed to a quality suitable for architectural visualisation, might run from one to several thousand euros depending on the detail required and the intended use.
For heritage or archival projects where extreme precision is required - submillimetre accuracy, full texture detail, deliverables that need to last decades - the cost is higher still, and rightly so. The brief and the stakes are different.
The most common mistake is under-specifying the brief. "A 3D model of our product" can mean twenty different things. What will it be used for, at what resolution, in what format? Those answers change the approach and the cost significantly.
Photogrammetry vs other 3D methods
vs manual 3D modelling: Manual modelling gives you clean, parametric geometry that is easy to modify - change the colour, adjust a dimension, swap a material. Photogrammetry gives you a realistic scan of the real thing, but the mesh is complex and harder to edit. For products that do not exist yet, or where you need variant flexibility, modelling wins. For capturing existing real-world objects accurately, photogrammetry wins.
vs LiDAR scanning: LiDAR (laser scanning) is more precise, more expensive, and produces cleaner point clouds - it is the professional standard for surveying and large infrastructure. Photogrammetry is significantly cheaper, works with standard cameras (including modern smartphones for lower-stakes captures), and produces textured models directly. For most commercial use cases that do not require survey-grade accuracy, photogrammetry is the better value.
vs photography: Photography captures one moment from one angle with a fixed depth of field. Photogrammetry captures the full three-dimensional reality. If you need to view something from any angle, place it in a 3D environment, or deliver an interactive experience, photogrammetry is the only option. If you just need a flat image for a specific use, photography is simpler and cheaper.
Is it right for your project?
Photogrammetry is worth considering when: the subject is a real physical object that needs to be captured accurately, when you need flexibility in how the 3D output is used, or when manual modelling would be prohibitively expensive relative to the complexity of the real thing.
It is the wrong choice when: you need clean CAD-like geometry for engineering purposes, when the subject does not yet exist physically, or when a single photograph would honestly serve the purpose just as well.
If you are unsure which side of that line your project falls on, get in touch. It is usually a short conversation to figure out whether photogrammetry is the right tool - and if it is not, what is.