Get3D Knowledge Center

03 LiDAR vs Photogrammetry

11 min read

LiDAR vs. Photogrammetry: Which Is Better for 3D Mapping?

Quick Answer

LiDAR and photogrammetry both create 3D data, but they begin with different evidence. LiDAR measures distance with laser pulses, making it strong for reliable geometry in low-light, low-texture, indoor, and vegetated environments. Photogrammetry derives 3D geometry from overlapping photographs, making it strong for realistic color, texture, and efficient coverage of visible surfaces.

For most projects, neither is universally “better.”

  • Choose photogrammetry when a realistic visual model, orthophoto, or broad-area capture is the priority.
  • Choose LiDAR when dependable geometry is the priority, especially in interiors, under partial vegetation, or around surfaces that are difficult to match in images.
  • Choose a hybrid workflow when the project needs both trustworthy geometry and a photorealistic surface model.

LiDAR vs Photogrammetry at a Glance

QuestionLiDARPhotogrammetry
How is 3D data created?A sensor emits laser pulses and measures the return distance.Software matches the same visual features across overlapping images.
Primary strengthDirect geometric measurement.Realistic color and texture at efficient area coverage.
Typical outputPoint cloud, often with intensity and RGB color.Point cloud, textured mesh, orthophoto, DSM/DTM, or 3D Gaussian Splatting.
Color and visual detailRequires a camera or separate imagery for rich color.Captures native imagery, so it is usually stronger for visual presentation.
Low-texture or dark surfacesOften remains usable because it does not depend on image texture or daylight.Can struggle where there are few stable visual features or insufficient light.
VegetationMultiple returns can reveal some terrain below canopy, depending on canopy density and survey design.Usually records the visible canopy rather than the ground beneath it.
Indoor and GNSS-denied areasWell suited to mobile or static scanning, often using SLAM.Possible, but coverage and lighting must be planned carefully.
Large-area city capturePowerful but can add sensor and processing cost.Efficient for roofs, terrain, facades, and visual city models, especially from oblique imagery.
Best role in a hybrid projectGeometric reference and coverage of difficult spaces.Surface texture, wide-area context, and visually complete delivery.

The table is a starting point, not a specification. A well-planned photogrammetry project can produce excellent geometry, and a poorly planned LiDAR project can still miss surfaces, drift, or lack the context needed for a usable model.

How LiDAR Creates 3D Data

LiDAR, short for Light Detection and Ranging, measures distance by sending laser pulses toward a surface and recording the time or phase of the return. Each return becomes a point with a position in three-dimensional space. A single scan produces a point cloud; many overlapping scans can be registered into a larger spatial model.

LiDAR can be collected from several platforms:

  • Terrestrial laser scanners for fixed, high-detail capture of buildings, industrial sites, and heritage assets.
  • Mobile or handheld scanners for walking routes through streets, interiors, tunnels, and complex facilities.
  • Drone or aircraft LiDAR for corridors, terrain, and larger sites where ground access is limited.

Because LiDAR measures range directly, it does not need a wall, road, or pipe to have a distinctive visual pattern. This is why it is useful in dark tunnels, on uniformly painted surfaces, and in scenes where image matching is unreliable. It does not, however, automatically produce a finished visual model. Point density, scan positions, control, registration quality, occlusion, and color capture still determine whether the data is fit for its intended use.

How Photogrammetry Creates 3D Data

Photogrammetry reconstructs a scene from overlapping images. Software identifies the same visual features in multiple photographs, estimates each camera position and orientation, and then derives the 3D location of visible surfaces through triangulation.

A typical workflow includes:

  1. Planning image coverage, viewing angles, lighting conditions, and control.
  2. Capturing overlapping imagery from a drone, aircraft, vehicle, handheld camera, or satellite.
  3. Solving camera positions through aerial triangulation or structure-from-motion.
  4. Producing dense geometry, a point cloud, mesh, orthophoto, elevation product, or 3DGS scene.
  5. Checking completeness, alignment, texture quality, and the intended delivery format.

For urban work, oblique photogrammetry adds angled views to the vertical camera view. These perspectives give the reconstruction pipeline the evidence it needs to model facades, roofs, streets, and other vertical structure in one broad-area capture. Since the source material is imagery, photogrammetry also carries the color and visual context that makes a model easy for non-specialists to understand.

The Differences That Matter in a Real Project

1. Geometry and Measurement

LiDAR starts with direct range measurements. That makes it a natural choice where the geometry itself is the primary evidence: deformation checks, as-built documentation, interior conditions, corridors, or terrain under partial vegetation.

Photogrammetry derives geometry indirectly from image overlap. With appropriate camera calibration, image coverage, control, and independent validation, it can deliver highly useful and accurate 3D data. But it is more sensitive to the quality of the imagery and to whether the surface contains enough stable visual detail to match.

The practical question is not simply, “Which technology is more accurate?” It is: what measurement tolerance, coordinate reference, validation method, and audit trail does this deliverable require? Those requirements should be set before data collection, not inferred from a good-looking model after the fact.

2. Visual Realism and Context

Photogrammetry usually has the advantage when a project needs a model that people can immediately read as the real world. It preserves facade materials, pavement markings, roof color, vegetation, signs, and the visual cues that make a city model useful for communication, planning review, and public-facing experiences.

LiDAR can be colorized by pairing it with imagery, but its native output is a geometric point cloud. That is often ideal for technical inspection, while a textured photogrammetric mesh or 3DGS scene is usually easier for broader audiences to explore.

3. Difficult Environments and Surface Types

Every capture method has blind spots.

Photogrammetry becomes more difficult when a scene has poor lighting, repeated patterns, transparent or reflective glass, moving objects, water, or broad surfaces with little visual texture. Good mission planning, multiple viewing angles, and careful processing reduce these risks, but they do not eliminate them.

LiDAR is less dependent on daylight and image texture. It can therefore provide valuable geometric evidence in many of those environments. It has its own limits: reflective materials, grazing scan angles, occlusion, sparse scan paths, and poor registration can create gaps or noise. A laser cannot measure a surface it never reaches.

4. Vegetation, Interiors, and Occlusions

For terrain modelling in wooded areas, airborne LiDAR is often preferred because some laser pulses can pass through gaps in the canopy and return from the ground. The result depends on vegetation density, season, sensor settings, flight design, and ground classification; LiDAR does not literally see through every tree.

For interiors, underground structures, and narrow urban spaces, ground-based LiDAR or LiDAR SLAM can capture geometry that an aerial camera cannot see. Photogrammetry remains valuable around those spaces for color, exterior context, and areas with strong image coverage.

5. Coverage, Processing, and Cost

Photogrammetry is often the efficient first choice for large, visible areas because a camera system can collect dense imagery of roofs, terrain, and facades in a planned flight. It can generate several downstream products from the same image block: orthophotos, meshes, point clouds, elevation models, and visual scenes.

LiDAR can reduce field revisits where image-based capture would leave geometric gaps, but it adds sensor, processing, and quality-control considerations. The cost comparison is therefore not just an equipment comparison. It includes field time, access constraints, rework risk, downstream requirements, and whether the same data must support both technical analysis and visual communication.

Which Should You Choose for 3D Mapping?

Project needStarting pointWhy
A visual city model, planning presentation, or public web experiencePhotogrammetryIt provides wide coverage and the color detail that makes the scene understandable.
A terrain model under partial tree canopyLiDARMultiple returns can improve the chance of recovering ground observations.
An interior, tunnel, station, or underground utility spaceGround LiDAR / LiDAR SLAMIt captures geometry where aerial visibility and GNSS are limited.
A building exterior with complex facadesOblique photogrammetry, with LiDAR where neededMulti-angle imagery provides rich visual surfaces; LiDAR can support challenging geometry.
Engineering review or an as-built condition that must be checked independentlyLiDAR or a validated hybrid workflowThe method should be driven by the specified tolerance and verification process.
A fast, repeated construction or site-progress recordPhotogrammetry, with targeted LiDAR for critical areasImagery is efficient to repeat; LiDAR can be reserved for geometry that needs a stronger reference.

The most useful choice is often not a sensor choice. It is a deliverable choice. Decide what stakeholders need to see, measure, edit, publish, or maintain, then plan the capture method that creates evidence for those needs.

When a Hybrid LiDAR and Photogrammetry Workflow Is the Better Answer

Hybrid capture is appropriate when one technology’s strength addresses the other’s weakness. Three patterns are especially common.

LiDAR for Geometry, Photogrammetry for Surface Detail

LiDAR supplies robust geometry or a reference point cloud, while photogrammetry adds the textures and visual context required for a realistic mesh or 3DGS scene. This pattern is useful for complex buildings, infrastructure, and communication-oriented digital twins.

Aerial Imagery for Coverage, Ground LiDAR for Blind Spots

Oblique aerial imagery efficiently captures rooftops, streets, terrain, and building exteriors over a broad area. Ground LiDAR or SLAM fills the areas that aircraft or drones cannot see well: street-level detail, interiors, tunnels, underpasses, and dense urban canyons.

Photogrammetry as the Baseline, LiDAR as Targeted Risk Control

Not every square metre requires LiDAR. A pragmatic workflow uses photogrammetry for the main model and sends LiDAR only to locations where the project cannot tolerate missing geometry or uncertain measurements. This can avoid both over-scanning and expensive rework.

A hybrid workflow still needs discipline. The datasets must share a coordinate reference, overlap in enough common areas to validate alignment, and be captured within a time window that does not introduce avoidable change between the two sources.

Four Questions to Ask Before Choosing

  1. Is the primary outcome visual, measurable, or both? A photorealistic walkthrough and an engineering measurement deliverable are not the same product.
  2. What parts of the site are hard to see? List canopy, interiors, tunnels, water, glass, narrow streets, and restricted-access areas before selecting a sensor.
  3. What must be independently validated? Define the control, checkpoints, tolerances, and acceptance method before acquisition.
  4. Will the model be updated or reused? A one-time presentation model and a living asset-management model justify different capture and delivery decisions.

These questions turn an abstract technology comparison into a defensible project plan.

Where Get3D Fits

Get3D supports the complementary roles of imagery and ground scanning in one reconstruction workflow. The 5-in-1 Camera captures multi-angle aerial imagery for broad, visually detailed city models. The R200 captures ground-level geometry in spaces that aerial capture cannot fully reach. Get3D Mapper then aligns and processes those sources into a unified output for mesh, point-cloud, or visualization workflows.

The advantage is not that every project must use both sensors. It is that teams can use the right evidence for each part of a project without treating aerial imagery and ground scanning as disconnected workflows.

Frequently Asked Questions

Is LiDAR more accurate than photogrammetry?

LiDAR provides direct range measurements, which is a major advantage for geometric work. But accuracy is a property of the complete workflow, not the sensor label. Calibration, control, mission design, overlap, registration, coordinate handling, and independent checkpoints all affect the result. A project should specify the required tolerance and validation method before choosing a technology.

Can drone photogrammetry replace LiDAR?

For visible, well-textured outdoor surfaces, drone photogrammetry can be an efficient and highly capable option. It is not a universal replacement for LiDAR in dense vegetation, dark interiors, tunnels, or geometry-critical work. The two methods are often complementary.

Can photogrammetry create a point cloud?

Yes. After the camera positions are solved, dense image matching can generate a colored point cloud. The same image set can also be used to create an orthophoto, elevation products, a textured mesh, or a 3DGS scene, depending on the workflow and delivery need.

Does LiDAR see through trees?

LiDAR does not see through solid vegetation. In an airborne survey, some pulses can pass through gaps in the canopy and return from the ground, which makes it useful for terrain modelling. Results depend on canopy density, survey conditions, and classification quality.

Source: Get3D Knowledge Center

Graph View