Capturing physical objects as printable 3D models — for reverse engineering, custom fit parts, replica creation, or design iteration on existing geometry — requires either structured-light scanning hardware or photogrammetry software. According to All3DP's 3D scanner overview, the accessible price range for structured-light scanning has dropped dramatically since 2020, while photogrammetry software has become free and desktop-capable enough to produce print-quality results from smartphone photographs. Both paths have distinct accuracy profiles, cost structures, and use-case fits that makers should understand before investing time or money.

Structured Light vs Photogrammetry: Core Differences

Structured-light scanning projects a known light pattern (typically infrared dots or lines) onto the subject and uses a camera to measure how the pattern deforms, calculating precise 3D coordinates from the deformation geometry. The technique is fast (capturing a full scan in seconds), works in a single pass, and produces point clouds that are accurately scaled and oriented in real space. Photogrammetry takes a different approach: algorithms analyze dozens to hundreds of photographs of the subject from different angles, detecting common feature points across images and triangulating their 3D positions from the known camera positions. Photogrammetry is slower (an hour or more for a complete capture-and-process workflow) but requires only a camera (including a smartphone), costs nothing in hardware, and can produce excellent surface texture capture alongside geometry. The practical distinction for makers is that structured-light scanning is better for capturing precise dimensions of mechanical parts and engineering geometry, while photogrammetry excels at capturing organic shapes, textures, and subjects too large to place on a scanning turntable — architectural features, terrain, full human figures.

Revopoint and Entry-Level Structured-Light Scanners

Revopoint has become the dominant brand in the entry-level structured-light scanner market, with their RANGE 2, MINI 2, and MIRACO models covering different size and accuracy tiers. The RANGE 2 targets medium-to-large subjects (50cm to 3m) with accuracy around ±0.1mm per frame; the MINI 2 is designed for smaller objects (5cm to 50cm) with tighter accuracy (±0.02mm) and finer detail capture. Both use infrared structured light with dual cameras and produce colored point clouds with surface texture when operated in color mode. Scanning workflow with Revopoint hardware involves either handheld scanning (moving the device around the object) or turntable scanning (rotating the object while the scanner is fixed) — turntable scanning with a motorized platform produces cleaner results with fewer tracking losses. Raw scan outputs require post-processing: the Revopoint RevoScan software handles point cloud cleaning, mesh generation, and export to STL or OBJ. The primary limitation of current entry-level structured-light scanners is struggle with reflective or dark surfaces — matte black objects and specular metal require spray coating with scanning spray (a removable white powder) before scanning for acceptable results.

Meshroom and Free Photogrammetry

Meshroom is a free, open-source photogrammetry application from AliceVision that produces print-quality 3D models from standard photographs on Windows, Linux, and macOS. The workflow: capture 40 to 150 overlapping photographs (minimum 60 percent frame overlap), import into Meshroom, and run the automatic pipeline covering feature detection, camera position reconstruction, dense point cloud generation, and mesh creation. Processing takes 30 minutes to several hours depending on image count and GPU capability. Meshroom excels on textured, matte surfaces in consistent lighting; smooth or reflective objects produce poor results because feature-matching algorithms cannot identify common points between near-identical-looking frames. Key tips: shoot in diffuse lighting to minimize specular highlights; include a known-dimension scale reference for accurate sizing; maintain consistent aperture for uniform depth of field across the capture.

Polycam and Mobile Scanning

Polycam is a smartphone application (iOS and Android) that uses photogrammetry and, on supported iPhone models, LiDAR depth sensing to generate 3D models from phone cameras. The LiDAR-based mode (available on iPhone 12 Pro and later) produces fast, room-scale scans suitable for architecture and interior spaces but with lower accuracy than structured-light scanning for small objects. The photogrammetry mode — available on all supported phones — produces quality comparable to Meshroom for well-lit textured objects when sufficient photos are captured. Polycam's primary advantages are convenience and processing speed: models are processed on cloud servers rather than local hardware, returning results in minutes regardless of local computing power. The trade-offs are ongoing subscription cost (the free tier allows limited exports, with full STL/OBJ export requiring a paid subscription) and the cloud dependency that means results require internet connectivity and involve uploading image data to Polycam's servers. For quick field scans, Polycam is highly effective; for production workflows where data privacy, offline operation, or cost-at-scale matter, Meshroom running locally is the more sustainable choice.

Use Cases and Realistic Expectations

Match the scanning method to the workflow. Structured-light is right for capturing precise mechanical geometry — brackets, housings, connectors — where dimensional accuracy matters. Photogrammetry suits organic shapes, costumes, terrain, and architectural elements too large for a turntable. Either way, expect post-processing: raw scan outputs require cleanup in Meshmixer or Blender to repair holes, smooth noise, and close the mesh. Plan for 20 to 60 minutes of mesh repair before a scan is print-ready. Accuracy expectations should match the method: Revopoint MINI 2 achieves ±0.05mm on prepared subjects; smartphone photogrammetry realistically delivers ±0.5 to 1mm on objects under 30cm. For engineering tolerances, use structured-light. For replica and organic work, photogrammetry accuracy is typically sufficient.

What It Means for Makers

3D scanning bridges the physical-to-digital gap without requiring CAD modeling skills — capturing a complex object and reproducing or adapting it faster than any manual reconstruction workflow. The practical barrier is no longer cost or complexity but workflow knowledge: which method to use, how to prepare the subject, and what post-processing the raw output requires. Start with photogrammetry using a smartphone and Meshroom — zero additional hardware cost — on a well-textured matte object, and evaluate whether its accuracy meets your needs before investing in dedicated scanning hardware.

Sources