Good photogrammetry of monuments comes down to three disciplines: even lighting, complete overlapping coverage at multiple heights, and reliable real-world scale from surveyed control points. Shoot on an overcast day or in soft light to avoid moving shadows, achieve 60–80% overlap on every facade with orbits at several heights (often adding a drone for the upper structure), and place distributed scale bars or surveyed ground control points so the finished model is measurable and defensible. A consistent, documented method is what keeps a whole collection of monument captures comparable.

When should I photograph a monument?

Lighting is the single biggest variable you control on a monument. Aim for:

• Overcast skies — a giant softbox; flat, even light with minimal shadow.
• Early morning or late afternoon if you must shoot in sun, but expect to work fast as shadows move.
• Avoid harsh midday sun — deep shadows on one shot become bright on the next as the sun shifts, and the matcher treats that inconsistency as noise.

Consistent illumination across the whole capture session is the goal; if it takes long enough that the light changes dramatically, the model quality drops.

How do I cover a large structure completely?

Coverage is about systematic orbits, not a count of photos. Plan it as layers:

Monument capture plan (per facade, repeat around the whole structure):
  Layer 1  ground orbit       ~ every 5-10 deg, camera level
  Layer 2  raised orbit       monopod/pole at 3-4m, angled up
  Layer 3  aerial orbit       drone, nadir + oblique passes
  Detail   close-ups          carvings, inscriptions, joints (>=80% overlap)
  Ties     corner shots       overlapping two facades to bind them
Overlap target: 60-80% between consecutive frames everywhere.

The corner and tie shots matter enormously: they are what stops the four facades aligning as four disconnected slabs. For tall monuments a drone is not a luxury — the roofline, cornices, and tops of buttresses are invisible from the ground.

How do I get accurate, real-world scale?

This is where monument work diverges from small-object photogrammetry. A single ground-level scale bar cannot reliably scale a 20-metre tower. Instead:

Method	Accuracy on large monuments	Notes
Surveyed GCPs (total station / RTK GPS)	High	Scale + georeferencing + error check
Multiple distributed scale bars	Moderate	Better than one; spread them in height
Single scale bar	Poor for big structures	Fine for small details only

Place ground control points (clearly marked, surveyed targets) distributed around and up the structure, measure them, and enter the coordinates in your software. They give scale, real-world position, and — because you can hold some back as checkpoints — a way to report actual accuracy rather than just claiming it.

How do I handle access limits and occlusion?

Real monuments are rarely fully accessible. You will hit railings, height limits, and protected areas. The honest approach:

1. Capture everything you can reach with complete overlapping orbits.
2. Raise the camera on a pole or monopod to claw back some height.
3. Use a drone where permitted for the rest.
4. Document the gaps. Note in your paradata which surfaces are incomplete or occluded, so a researcher does not mistake missing geometry for a measured flat.

An incomplete-but-honest model beats a hole-filled model that hides where the data ran out.

What is a repeatable monument workflow?

Consistency across a collection matters more than perfection on one capture. A defensible standard sequence:

1. Plan — recce the site, check light, permissions, and drone rules.
2. Control — survey and mark GCPs before shooting.
3. Capture — layered orbits, corners tied, detail close-ups, consistent exposure (manual mode, fixed focus per orbit).
4. Process — align, place GCPs, build dense cloud and mesh, report checkpoint residuals.
5. Document — paradata: operator, date, kit, GCP method, accuracy, occlusion notes.
6. Archive — open formats, source photos, checksums, the same way every time.

Write this down as your collection standard so every monument is captured comparably, by anyone on the team.

What are the common pitfalls?

The recurring mistakes on monuments: shooting in moving sun; forgetting tie shots at corners (causing split alignment); relying on one scale bar; skipping the upper surfaces because they are awkward; and failing to record accuracy. Each is avoidable with the checklist above, and each quietly undermines whether the resulting model can be trusted for measurement or conservation planning.

Key Takeaways

• Shoot monuments in flat, even light (overcast or soft) to avoid moving shadows.
• Cover every facade with 60–80% overlap in layered orbits at multiple heights.
• Tie facades together with overlapping corner shots to prevent split alignment.
• Use surveyed ground control points for scale, georeferencing, and accuracy checks on large structures.
• Add a drone for upper surfaces and rooflines you cannot reach from the ground.
• Document occlusion and gaps honestly rather than hiding them with hole-filling.
• Standardise the whole workflow so every monument in a collection is captured comparably.

Frequently Asked Questions

What is the best time of day to photograph a monument for photogrammetry?

Overcast days or the soft light of early morning and late afternoon are ideal because they give flat, even illumination with no hard shadows. Harsh midday sun bakes in shadows that move between shots and ruin the texture.

How do I get scale on a large monument?

Place several scale bars or measure ground control points with a total station or RTK GPS, distributed around and up the structure. For tall monuments, surveyed control points give far more reliable scale than a single ground-level scale bar.

Do I need a drone for monument photogrammetry?

Not always, but for anything tall a drone captures the upper surfaces and roofline you cannot reach from the ground. Combine drone and ground photography into one project so the whole structure is covered and tied together.

How many photos does a monument need?

It scales with size and detail, often hundreds to low thousands. The rule is coverage, not count: 60-80% overlap on every surface, orbits at multiple heights, plus close-ups of detailed areas like carvings and inscriptions.

How do I handle a monument I can only walk around at ground level?

Shoot complete overlapping orbits at several heights using a monopod or pole to raise the camera, capture every facade and corner, and accept that occluded upper surfaces will be incomplete. Document the gaps in your paradata.

Should I use ground control points or just scale bars?

Use surveyed ground control points (GCPs) for any sizeable monument: they provide scale, real-world georeferencing, and a way to check accuracy. Scale bars alone are fine for small features but drift on large, tall structures.

Related reading

• How to capture photos for photogrammetry
• How to scale 3D models
• Photogrammetry vs LiDAR
• Documenting 3D model provenance
• Georeferencing historical maps