GenAnywhere Generate Now

Select a map area. Generate a 3D city model with AI.

No drone survey required. Turn map data into editable 3D buildings, terrain, water and vegetation for BIM, GIS and 3D workflows.

Generate now

Oblique photography vs site models: what each can show, GenAnywhere

One records a place as it looked; the other arranges a place as it could be. Teams need both, and rarely need them mixed.

Oblique photography and the site models drawn from map data are often discussed as if they were two grades of the same product. They are not. One is a photographic record turned into a mesh, the other is an editable arrangement built from map data. Knowing which question each answers prevents a review from arguing about the wrong evidence.

Aerial view of a hillside town with terraced streets and trees

1 Two records, two purposes

1.1 What does each produce?

An oblique capture produces a textured mesh. Cameras at several angles record façades and roofs, and photogrammetry reconstructs the surfaces into a model that can be viewed and measured.

A map-derived base produces an editable scene. GenAnywhere draws an area from map and satellite data, identifies footprints, estimates heights, infers building types and reads vegetation, then separates buildings, terrain, water and vegetation into layers.

One is an artefact of a flight, the other is a working model of an area. The difference shows up the moment someone tries to edit a wall. The two are often produced by different suppliers, which is another reason to name the job each one is doing.

1.2 Why does the difference matter?

Because the two carry different kinds of confidence. A capture is good evidence that a surface looked a certain way on a date. A map base is good evidence of arrangement, and weaker evidence of any particular surface.

Reviews fail when confidence is transferred. A photo-real façade invites people to trust the dimensions beside it, and a clean block model invites people to trust its edges. A presentation that says "the model shows" when it means "the capture shows" will be corrected in public.

Naming the kind of confidence at the start of a meeting saves an argument later. Photogrammetric societies such as ISPRS publish material on how these records are produced and where their limits sit.

2 What the capture shows

2.1 Surfaces and condition

The capture shows what a camera could see: materials, weathering, signage, the state of a roof, the position of a fence. For a condition survey or a heritage record, that is the substance of the work.

It also shows change. Two flights a year apart, viewed side by side, will show a new extension or a cleared yard more convincingly than a drawing revision. They also show what did not change, which matters when a claim about deterioration is being tested.

What it cannot show is what lies behind a surface. A rendered wall, a covered window and a false façade all read as flat planes.

2.2 Edges and awkward geometry

The capture preserves the parts a map generalises: the lean-to, the steps, the retaining wall, the gap between two buildings that is narrower than a drawing suggests.

That makes it useful for checking a site's real boundaries. A fence line that runs two metres inside the mapped parcel is the kind of detail that changes a scheme. Teams often fly a site for this reason alone, before any design work begins.

The same detail makes the mesh heavy. Awkward geometry costs file size, and a capture that covers a district at façade detail will open slowly.

Aerial view of a riverside town beside a green floodplain

3 What the base shows

3.1 Arrangement and layers

The base shows how a place is put together: which blocks face which street, where the ground rises, how the water sits against the bank.

Because GenAnywhere keeps buildings, terrain, water and vegetation on separate layers, a team can hide the trees to study the street pattern, or hide the buildings to study the ground. A base that keeps those layers apart can be read by three specialists in one meeting without anyone losing the thread.

That separation is what makes the base useful for a proposal. A planner can replace a building, move a block or change a height without rebuilding the scene.

3.2 Alternatives and comparison

The base is a place to test options. Two massing schemes can sit in the same frame, and a reviewer can compare them from the same viewpoint. Name each option scene by version, so the approved one cannot be confused with the one that was abandoned.

It also supports a scale of study that no flight covers. A district review can extend across ground that was never photographed, at a consistent level of detail.

Aerial view of a city beside a lake and wooded hills

What it does not support is a claim about condition. A grey volume says nothing about the state of the building it stands for. The export is a general GLB, so the arrangement can travel into a BIM or GIS workflow while the capture stays a reference.

4 Where they disagree

4.1 Which source wins?

The source that carries responsibility for the decision. If the question is whether a façade is intact, the capture wins. If the question is whether a new block fits, the map base wins.

Where both speak to the same point, the more controlled source should govern, and the team should say which one that is.

Write the choice down. A note that says "dimensions follow the survey, arrangement follows the base" prevents a later reader from assuming the opposite. A short statement at the front of the pack settles most of those arguments before they start.

4.2 How do you mark the difference?

Style the capture and the base differently. Texture for what was photographed, flat or hatched material for what was modelled.

Label proposed elements. A new volume in a captured scene should not look like a photograph, or the room will treat it as existing.

Keep the two as separate files where possible. A merged scene is convenient to present and hard to audit. That trade is worth naming in the note rather than discovering in a review.

5 Choosing between them

5.1 When do you fly?

Fly when the question depends on appearance, condition or recent change. A façade survey, a heritage record, a completed scheme or a construction sequence all justify a flight.

Fly when the site is irregular. Photogrammetry records the awkward parts that a map will smooth away.

Fly when the surroundings matter at surface level, such as a proposal beside a listed building where the texture of the neighbour is part of the argument. Fly when the record has to survive a question about what the site looked like on a particular day.

5.2 When do you draw?

Draw when the question is about arrangement. A blockout, a density test, a route study or a massing comparison needs volumes that can move.

Draw when the area is larger than a flight. A district or a corridor is easier to cover with a map base than with repeated captures.

Draw when the site is not yet built. A capture records what exists, and a proposal needs a base that can hold what does not. Draw when the team needs to try three arrangements before lunch.

6 Using them together

6.1 What does a combined scene look like?

The capture sits as the record of existing conditions, and the map base carries the proposal and the wider context. The two are aligned, labelled and kept distinguishable.

A reviewer can then see what is there and what is proposed in one frame, without confusing the two. The discipline is in the labels, not in the software.

That is the practical version of "existing and proposed" that most planning conversations need.

6.2 What does combining cost?

An alignment check, a labelling convention and a note. The technical work is modest; the discipline is what takes effort.

Time goes into deciding which source governs which dimension, and into keeping the files separate enough to audit.

Skipping that work produces a scene that looks finished and cannot answer a question about where a number came from. Budget the alignment check rather than discovering it during a presentation.

6.3 When do you stop adding?

Stop when the decision has an answer the team can state in one sentence. Extra texture does not make a massing judgement more right.

Stop when the file becomes slow enough that people avoid it. A scene opened once is not a working tool.

Stop when the next question needs a different discipline. Hand over the record rather than stretching it. A record handed on is more valuable than a scene kept perfect.

Draw the area in the GenAnywhere app and keep the capture beside it as a dated reference.

7 FAQ

FAQ-1 Can GenAnywhere turn photos into a model?

No. The platform builds a site model from map and satellite data, and angled images are processed in photogrammetry software. The two results can be used side by side, but they are produced by different methods. If a team needs a mesh from photographs, that work happens before the scene reaches the platform.

FAQ-2 Which is more useful for a planning application?

It depends on the question. A planning conversation usually needs arrangement and context, which the map base supplies, and may need surface evidence for a heritage or design point, which the capture supplies. Most teams use both, and a pack that labels each source is easier to defend than one that leans on a single record.

FAQ-3 Can I edit the capture in GenAnywhere?

No. A captured mesh is not editable in the platform. GenAnywhere edits its own model, where buildings, terrain and vegetation are separate layers and can be replaced or moved. The capture stays a reference file, and edits happen on the model, where the parts are separate. A team that needs both keeps two files and one note explaining how they relate.

FAQ-4 How do I explain the difference to a client?

Describe one as a photograph you can walk through and the other as a model you can rearrange. Then say which questions each answers. Clients usually grasp the distinction quickly once it is put in those terms.

See how a map area becomes an editable site model in the GenAnywhere workflow.

Content and technical notes

This article was prepared by the GenAnywhere team with AI-assisted drafting and human review.

Product features, formats, and availability change as the platform evolves; refer to the official GenAnywhere website for current details.

For technical questions or corrections, contact support@genanywhere.com.

Last reviewed: 2026-11-06.