Now available: 3D Model Object Tracking

Eva Triantafillopoulou

You spoke. We listened. Work with the new Wikitude SDK 9.2 to create high-performing Object Tracking AR experiences with 3D model files.

Wikitude products are being constantly perfected, adapted, corrected, and updated every couple of months to ensure our customers have access to the market’s latest and greatest AR features and top-performing AR tools.

This Wikitude SDK 9.2 release introduces highly requested new features and functions, and several product improvements and updates:

Are you a subscription client? Download Wikitude SDK 9.2 to update your platform.

Wikitude SDK Expert Edition 9.2

Wikitude SDK 9.2 Expert Edition features at a glance:

• 3D Model Object Tracking (Beta)
• Alignment Initializer
• High-end Image Tracking (single & multiple)
• Cylinder Tracking (single & multiple)
• Cloud Recognition v.2
• Point cloud rendering in Unity Editor
  General improvements and additional features: edit color and point size
• Update OpenSSL
  Encryption library update brings more communication security
• Multiple Trackers (image + object + positional tracking)
  Enhance images, objects, and the world in one unique AR experience.

New and improved features in Expert Edition 9.2

3D Model Object Tracking (beta)

Wikitude Object Tracking technology is the AR feature of choice when digital content needs to be seamlessly superimposed and attached to real-world objects.

Today, Wikitude is proud to introduce the beta version of a new object target input method using 3D models. This feature enables better recognition and tracking and incorporates a greater range of recognizable and trackable physical objects.

As many of you may already know, a Wikitude Object Tracking AR experience is triggered when the user’s device finds an equivalent target reference in the live camera viewfinder. This target reference is a digital pre-recorded map of the physical object called object target(.wto). The object target reference map, up until today, could be created with images as the source material.

With Wikitude SDK 9.2, beta testers now have the additional possibility of using 3D models as the source material for the object target reference map creation process.

3D Model Object Tracking benefits at a glance:

• More object target input methods available: images and 3D models can be used as source materials
• Input source materials are interchangeable and combinable: work with solely images, solely 3D models, or a combination of the two
• Greater range of recognizable and trackable physical objects: using 3D models as input is ideal for feature-poor, reflective, and low-textured objects
• Improved recognition and tracking performance: 3D models have the potential to expand the target map and improve object tracking performance

Access our augmented reality tracking guide to learn more.

Images or 3D models: the choice is yours.

By adding the option to work with 3D models as input for creating object references, Wikitude now offers the broadest choice when it comes to working with Object Targets.

From a developer perspective, both approaches (image-based and 3D model) have their justification. Advantages of both approaches:

Start working with 3D Model-based Object Tracking: beta testing program

Are you already working with 3D-model projects and have experience with AR? See the potential for an interesting 3D model based Object Tracking AR use case? Contact us for a chance to be part of the beta testing team. Selected partners will receive a free individually customized Object Target reference map for testing.

Alignment Initializer

The Wikitude AR SDK offers an Automatic Initialization process, meaning a user can recognize and track objects (and images) by freely pointing a device from any angle and distance.

SDK 9.2 introduces an alternative process called Alignment Initialization that guides the user towards a specific (usually ideal) viewpoint when pointing a device to recognize an object.

Benefits of the new Alignment Initializer object recognition function:

• Can be used for objects that are hard to recognize automatically: objects that have no clear texture or texture is unreliable. An unreliable texture could be the case that an object has different colors or that the areas keep changing (e.g. mud, stickers,…).

• Guides the user to the ideal viewpoint (perspective) from which the object can be recognized and tracking can be started: this viewing guide helps to communicate what the user is expected to do. The actual UI implementation of the Alignment Initializer can be freely chosen by the developer and can range from showing an outline to a rendered 3D view that can be manipulated by the user.

Testing Alignment Initializer: sample app

Download Wikitude SDK Expert Edition 9.2 to test the new Alignment Initializer feature which contains an Alignment Initialization AR experience using a traditional traffic stop sign.

High-end Image Tracking (single & multiple)

The Wikitude team works closely with clients and thankfully welcomes all product feedback coming from our awesome AR community. Thanks to this fruitful exchange, we were able to implement considerable Image Tracking improvements in this release.

When working with Image Tracking in SDK 9.2 expect great tracking stability when the experience is submitted to short and quick hand movements during the AR scan (independent of ARKit/ARCore). When ARCore/ARKit is present, SDK 9.2 users can expect even more stabilized tracking for image targets.

Cylinder Tracking (single & multiple)

SDK 9.2 brings improved tracking stability for Cylinder targets when ARCore/ARKit is present.

Cloud Recognition v.2

The previous Wikitude release (SDK 9.1) introduced a new generation of online Image Tracking allowing users to generate image target archives up to 1100x faster with Cloud Recognition v2.0.

Since then, our technical team ran a series of tests and experiments to optimize the service even further:

Thanks to an adaptation of the data communication between the SDK and the server, image targets are now recognized from twice the distance using Cloud Recognition v2.0.

Bring your large image collection to life and work with the powerful features and upgraded performance of the new Wikitude Cloud Recognition v2.0.

Wikitude SDK Professional Edition 9.2

Wikitude SDK 9.2 Professional Edition features at a glance:

• High-end Image Tracking (single & multiple)
  Great tracking stability with short and quick hand movements during the AR scan
• Cloud Recognition v.2
  Image targets are now recognized from twice the distance
• Point cloud rendering in Unity Editor
  General improvements and additional features: edit color and point size, …
• Update OpenSSL
  Encryption library update brings more communication security
• Update Flutter 1.17
  Power your Flutter projects with Wikitude AR
• Geo AR Tracking
• Instant Tracking
• Object Tracking

Download Wikitude SDK 9.2

Active Wikitude SDK subscribers are entitled to the respective SDK version updates released throughout their term. Follow the links below to update your Wikitude SDK license:

• Wikitude AR SDK for Android
• Wikitude AR SDK for iOS
• Wikitude AR SDK for Windows
• Wikitude AR SDK for Unity
• Wikitude AR SDK for Cordova
• Wikitude AR SDK for Xamarin
• Wikitude AR SDK for Flutter
• Wikitude AR SDK for HoloLens

New to Wikitude? Download a free Wikitude trial version for testing purposes and contact our team to discuss upgrade possibilities.

To explore all SDK options, including smart glasses, plugins, and other dev tools, please access our download page:

Interested in creating an AR project of your own? Access our store to choose your package or contact our team to discuss your specific AR requirements in detail.