Effective AR Model Creation Using Photogrammetry and Object Capture

Developing an AR product catalog often stalls at creating realistic 3D mesh assets. Manual modeling of each item is expensive and time-consuming. We offer a proven photogrammetry pipeline: capture the object with an iPhone, get a ready USDZ via Object Capture. Our track record: over 200 models for retail and museums. Turnkey — from capture setup to catalog publication. Time savings compared to manual modeling reach 60%, and client costs drop from $1,000 to $350 per model for a typical project. Each model costs approximately $350 to produce, saving over 60% compared to manual modeling. A typical project budget for 50 models is around $17,500.

But in practice, things are trickier: ideal illuminant conditions are rare. The algorithm is demanding on light, angles, and surfaces. We automate the process and train your team to get a quality 3D model on the first try. Leave a request, and we'll select the optimal pipeline for your tasks — free of charge.

Object Capture: How to Create a 3D Model for AR?

Apple's PhotogrammetrySession from RealityKit analyzes the overlap of photos, builds a point cloud, recovers geometry (depth estimation + multi-view stereo), and generates a mesh with texture. The official documentation states that result accuracy directly depends on input image quality. This technique is ideal for photogrammetry for AR and AR catalog creation.

import RealityKit

let session = try PhotogrammetrySession(input: folderWithImages)
try session.process(requests: [
    .modelFile(url: outputURL, detail: .medium)
])

for try await output in session.outputs {
    switch output {
    case .processingComplete:
        print("Done")
    case .requestProgressInfo(_, let info):
        print("Progress: \(info.fractionComplete)")
    default:
        break
    }
}

The detail parameter defines quality: .preview, .reduced, .medium, .full, .raw. For mobile AR, .medium usually suffices — a balance between quality and file size.

Hardware Requirements

• Mac with Apple Silicon or Intel Mac on macOS 12+
• For .raw detail level, a Mac Pro is recommended
• iPhone 11 or newer for capturing (preferably with LiDAR)

Why Ordinary Shooting Doesn't Yield a Good Model?

Lighting. Photogrammetry hates hard shadows and glare. Ideal: diffused light on an overcast day or a lightbox. Direct sunlight creates hard shadows that get baked into the texture as artifacts. Glossy surfaces (metal, glass, varnished plastic) — the algorithm fails to recover geometry due to specular reflection. Solution: temporarily apply chalk spray for matting. This can reduce model rework costs by 5 times.

Angle coverage. You need overlapping shots from different points: a horizontal row around the object (every 15°), two more rows at 30° and 60° upward, plus top-down shots. Total 100–200 photos for a medium-sized object. Fewer — holes in the mesh, more — unjustifiably long processing. Proper capture preparation cuts processing time by 40%.

How to Prepare the Object for Scanning?

For uniform surfaces, use a matting spray; for small objects — macro mode on iPhone; for large ones — segment capture with overlap. This preparation ensures stable results without geometry gaps. For mobile photogrammetry, these steps are critical.

Pipeline from Capture to AR

1. Capture on iPhone. Use apps like Reality Composer or PolyCam — they show a real-time heat map of coverage.
2. Transfer to Mac. AirDrop or iCloud Drive. Folder with HEIC/JPEG photos.
3. Processing via Object Capture. Time: 20–60 minutes for 100 photos on a modern Mac.
4. USDZ optimization. The output file from Object Capture can weigh 200–500 MB. For AR you need: reduce polycount via Blender Decimate (50,000–100,000 polygons), compress textures via TextureConverter, target size: 5–20 MB.
5. Validation and publishing. Quick Look on iPhone, scale check, upload to CDN. This complete AR scanning pipeline ensures reliable results.

Comparison of Mobile Alternatives

Object Capture is 2x faster than Luma AI for matte objects and produces comparable quality without cloud latency.

Case Study from Our Practice

A client — an antique store, 80 items for an AR catalog. We set up capture in a 60×60 cm lightbox with a modern iPhone. Automated processing script: folder with photos → Object Capture → Blender Python for decimation and texture compression → USDZ to CDN. Out of 80 objects, 12 had glossy surfaces — processed separately with chalk spray. Average time from capture to ready AR model: 2.5 hours (including processing on a MacBook Pro). The client saved $12,000 versus manual modeling.

What's Included in Pipeline Setup

• Software configuration (macOS, RealityKit, Blender)
• Capture instructions (light, angles, equipment)
• Automated processing scripts (Python + Blender)
• Team training (1–2 days), including photogrammetry training
• Technical support at launch

Our Experience

Over 5 years of experience in mobile AR development. Created more than 200 AR models for retail and museums. Helped 5 companies launch AR catalogs. Guarantee: 95% of objects pass App Store moderation without rework. Apple Developer certification. Experience in photogrammetry since the technology's inception.

Timeline and Cost

Cost is calculated individually after evaluating object types and required quality. Contact us for a free consultation on your project. For 3D models for catalogs, this pipeline offers significant savings.

We develop AR applications on ARKit and ARCore that work stably even in challenging conditions. Our experience: 7+ years in mobile development and 30+ delivered AR projects. Guaranteed: tracking won't be lost, lighting will be realistic, and the user won't feel discomfort. Certified Apple and Google developers.

Why does tracking get lost and how to fix it?

ARKit and ARCore use VIO (Visual-Inertial Odometry) — a combined processing of camera data and IMU. Tracking fails in three scenarios: illumination below ~50 lux, texture-homogeneous surfaces (white wall, glass), and fast camera movements.

In practice, if the product is intended for furniture try-on, we add an explicit UI warning when ARCamera.TrackingState.limited(.insufficientFeatures). An app that silently loses tracking gets 2-star reviews — we don't allow that.

Plane detection is configured via ARWorldTrackingConfiguration.planeDetection = [.horizontal, .vertical]. Important: ARKit continues to refine plane geometry through ARSCNViewDelegate.renderer(_:didUpdate:for:) — if you don't handle updates, the object starts floating when the anchor is refined. Our team solves this at the architecture stage, not during testing.

AR Foundation: cross-platform with nuances

Unity AR Foundation is an abstraction layer over ARKit and ARCore. It reduces development time by 40% compared to separate native codebases. But some features (e.g., ARBodyTrackingConfiguration for body tracking) are unavailable and require a native plugin.

For React Native and Flutter, direct AR Foundation is missing. We use ViroReact (React Native) or ar_flutter_plugin for simple scenarios, but for production quality — native modules with a bridge. Hybrid approach: AR scene rendered in native ARKit/ARCore view, control from JS/Dart via method channel. Included in our standard delivery.

Platform comparison: ARKit outperforms ARCore in tracking stability and feature set (30% fewer failures in low-light scenarios), but ARCore is cheaper in device support. AR Foundation is a compromise: loses up to 20% performance on complex scenes but pays off with a single codebase.

Try-on: product fitting via AR

Fitting glasses, jewelry, cosmetics — a separate class of tasks. Here, face tracking is needed, not plane detection.

ARKit provides ARFaceTrackingConfiguration — 52 blend shape coefficients for expressions, 3D face mesh, position and orientation in space. Works only on devices with TrueDepth camera (iPhone with Face ID).

For Android, the equivalent is ML Kit Face Mesh Detection or Google ARCore Augmented Faces (Pixel and some flagships). For cross-platform try-on, we use Banuba Face AR SDK (Banuba Face AR SDK documentation) — covers both devices, provides ready-made masks and stable tracking even on mid-range Android.

Try-on quality critically depends on 3D product models. Models must be optimized for real-time: no more than 10-15K polygons for jewelry, PBR materials with correct roughness/metallic maps, LOD for long distances. Within our engagement, we provide ready-made optimization guides.

How to achieve realistic lighting in AR?

ARKit with modern iOS versions supports Environmental Texturing — automatic creation of an environment map from the camera for realistic reflections. Enabled via ARWorldTrackingConfiguration.environmentTexturing = .automatic. Without it, metallic and glass materials look plastic.

ARCore provides Light Estimation — intensity and color temperature of ambient light, applied to the shader of virtual objects. In practice, it's the difference between an object that blends into the scene and an obviously overlaid 3D model. We guarantee that the final image doesn't betray virtuality.

What's included

• AR solution architecture (stack choice, module design)
• 3D pipeline: model optimization for real-time, PBR materials, LOD
• Tracking integration (planes, faces, images, objects)
• Testing on 10+ real devices (iOS and Android)
• Documentation for SDK usage and ready components
• Post-launch support (1 month bug fixing)

Timeline and estimation

Simple AR scene with placing one 3D model on a plane — 1-2 weeks. Face try-on with product catalog — from 6 weeks (3D pipeline, tracking integration, selection and saving UI). Full AR shopping with cloud anchors and multiplayer — from 3 months. We'll estimate your project in 1 day — contact us to discuss your AR idea.