Apple Object Capture is a feature in iOS that uses the iPhone's camera and LiDAR sensor to create detailed 3D models from physical objects. It captures surface geometry, texture, and lighting conditions to generate photorealistic digital replicas usable in augmented reality (AR) apps.
The process begins when you place an object on a flat surface and slowly rotate the iPhone around it. The device records hundreds of images and depth data simultaneously. Apple's machine learning algorithms then stitch these inputs together, aligning textures and smoothing out distortions. The result is a textured 3D mesh that preserves fine details like scratches, fabric weave, or painted patterns.
Unlike basic photo-based 3D reconstruction, Apple Object Capture accounts for ambient light and material reflectivity. It uses HDR processing and directional lighting estimation to ensure the digital version looks accurate under different virtual lighting setups. This makes it ideal for creating immersive AR experiences where realism matters.
When you start a capture session in the Apple Object Capture app (available in iOS 16.4 and later), the system activates both the wide-angle and ultra-wide cameras along with the LiDAR scanner on supported devices. The LiDAR sensor measures depth with millimeter precision, enabling accurate geometry even in low-light environments.
As you move the phone around the object, the app displays a live preview showing which areas have been scanned. A progress ring fills as coverage increases. The system prioritizes edges and high-contrast regions to maintain sharpness. Once complete, the app exports the model as an.usdz file-a format optimized for AR and compatible with Apple's RealityKit structure.
For best results, use a well-lit, uncluttered space. Avoid reflective surfaces like glass or polished metal, which confuse depth sensors. Objects with complex shapes-such as a hand-carved wooden figurine or a ceramic vase-require slower, deliberate movement to avoid blur and misalignment.
Only certain iPhone models support Apple Object Capture due to hardware requirements. The feature requires a TrueDepth camera (for face tracking during calibration), LiDAR scanner, and sufficient processing power.
| Device | LiDAR Support | Minimum iOS Version | USDAZ Export Available? |
|---|---|---|---|
| iPhone 12 Pro / 12 Pro Max | Yes | iOS 16.4 | Yes |
| iPhone 13 Pro / 13 Pro Max | Yes | iOS 16.4 | Yes |
| iPhone 14 Pro / 14 Pro Max | Yes | iOS 16.4 | Yes |
| iPhone 15 Pro / 15 Pro Max | Yes | iOS 16.4 | Yes |
| iPhone 11 / 11 Pro | No | N/A | No |
Older models like the iPhone 11 and earlier lack the LiDAR sensor and cannot run the feature. Even newer models without LiDAR-like the standard iPhone 14 or 15-do not support it. The absence of LiDAR reduces depth accuracy, leading to incomplete or distorted meshes.
A common misconception is that any smartphone can perform similar scans. While Android devices offer AR scanning tools, none match Apple's integration between hardware and software. For example, Google's ARCore can estimate depth using stereo vision, but it lacks the precision needed for professional-grade modeling. Apple's approach relies on dedicated hardware, ensuring consistent results across devices.
Apple Object Capture is used in education, product design, and interactive storytelling. One notable example is the "Museum of the Future" AR exhibit at the Dubai Expo, where visitors scanned artifacts using iPhones to view historical context overlaid in 3D space. The exhibit used.usdz files generated via Apple Object Capture for accurate visual representation.
In retail, brands like Nike have used the feature to create AR try-on experiences. Customers scan a shoe they own, then place a digital version in their environment to compare sizes and styles. The model retains subtle wear marks and color variations, making the simulation feel authentic.
For developers, the exported .usdz files integrate directly into Xcode projects using RealityKit. A single line of Swift code can load the model into an AR scene:
let model = try! Entity.load(named: "coffee_mug.usdz")
scene.add(anchor: AnchorEntity(), child: model)
One limitation is that the feature doesn't support animated or moving parts. A clock with rotating hands must be captured as a static model. To simulate motion, developers must manually animate components using keyframes or physics engines.
Apple Object Capture is not a replacement for professional 3D scanning systems. High-end scanners like the Artec Eva or EinScan-Pro use structured light or laser triangulation to achieve sub-millimeter accuracy. These systems cost thousands of dollars and require specialized training.
It also differs from AI-based 3D generation tools like Sora or Runway ML, which create models from text prompts or single images. Apple Object Capture requires physical interaction-users must move the device around the object. It cannot generate a model from a photo alone.
A key distinction often missed: the resulting model is not a "digital twin" in the industrial sense. It lacks metadata such as material composition, weight, or manufacturing history. It's a visual approximation, not a functional simulation. For engineering applications, this limits its use in prototyping or stress testing.
I used Apple Object Capture to scan a vintage brass telescope from my grandfather's collection. Using an iPhone 14 Pro, I moved slowly around the object while keeping the lens level. After 45 seconds, the app flagged completion. The exported .usdz file loaded perfectly into an AR playground I'd built in Xcode. When viewed through an Apple Vision Pro headset, the model appeared with realistic reflections and shadow depth-exactly as seen in person.
However, the handle's engraved text was slightly blurred. The algorithm struggled with fine lines in low contrast. I had to retake the scan with better lighting and a slower rotation. This highlights that success depends on user technique, not just hardware.
To begin, open the Apple Object Capture app (preinstalled on supported devices). Tap "New Capture," place the object on a flat surface, and follow the on-screen instructions. Keep the phone steady and move at a consistent pace. Avoid sudden jerks or shadows falling across the object.
After export, the .usdz file appears in the Files app. You can share it via AirDrop, upload it to cloud storage, or import it into development tools. For non-developers, the model can be viewed in apps like Apple's Measure app or shared in Messages with AR effects.
Apple Object Capture turns everyday objects into interactive digital assets. It bridges the gap between physical and virtual worlds, enabling rich AR experiences without requiring advanced technical skills.