© 2021 Visometry GmbH all rights reserved.
VisionLib Release 20.3.1 is now available at visionlib.com.
This general update of VisionLib adds official support of HoloLens 2, by providing ARM and ARM64 builds for the SDK. It also brings many smaller & bigger updates that ease usage and workflows.
Besides, there are many bugfixes that improve performance and stability. Read on.
Have you ever seen objects tracked on Hololens through the glasses? Here you go: VisionLib’s Model Tracking on HoloLens 2, officially supported in the new release. It enables to precisely localize and track objects on the #XR glasses.
To show the results, we’ve filmed through HoloLens 2 glasses with a smartphone. Slight augmentation offsets are caused by this capturing setup and not by the tracking.
HoloLens is great for AR-enhanced maintenance or repair overhaul tasks, because you have your hands free, and still get an augmented information layer while working. Registering the objects and precisely positioning information on them is key for it.
We’ve introduced a more central VLTrackingConfiguration object, where you can reference your tracking configuration and start tracking from.
We also updated the example scenes, with better descriptions and links to the documentation. And we added an example scene to enable image sequence recording on HoloLens.
We’ve improved debug information. There is now a full screen debug image, live updates while tweaking tracking parameters during runtime in Unity, and better notifications.
The new JSON schema brings a faster editing of tracking configuration files with auto-completion and suggestions for supported editors, because it helps finding parameters quicker and setting their values correctly.
Its included in all configuration files starting with this release, and you can add it to exsisting ones using this line of code in your (.vl) tracking configuration:
We’ve also updated our AutoInit (beta) feature, which enables to initialize and track objects 360 degrees from literally any angle. It liberates not only from Model Tracking’s (single) initialization pose, but is essential to multi model tracking as well. See how this feature makes it possible to measure the distance between two objects in the video above.
With this release official support for Unity versions below 2018.4 LTS has now ended. There are also some changes coming that might affect or break existing projects. For a safe update process, please consult the Migration Guide. An overview on all updates – including changes, enhancements and bugfixes to functionality & interfaces for Unity and native platforms – is given in the full release notes.
We’d also like to give a heads up: with the next release we are going to introduce even more fundamental breaking changes in order to make workflow and developments faster and easier. Stay tuned.
The Corona pandemic struck the world rather quickly and, depending on where you work and live, has led to strong social and economic restrictions, deeply affecting our everyday lives. While our team is well working from home, Corona affects conferences and shows and forces organizers to cancel or postpone essential get-togethers, which the XR community and we had planned for.
In the meantime we encourage you to follow us on LinkedIn, Twitter, and to have a look on our YouTube channel: we share insights into current developments, present new or upcoming features, and share exciting work, partners & customers have created using VisionLib.
Stay healthy and stay tuned. We keep you posted.
Augmented Reality is in the midst of today’s technology portfolio. But when it comes to complex AR cases, working with computer vision tracking can become intense. Yet, we get you quickly started with VisionLib:
Check our revised Example Scenes in the Unity SDK. Our Using VisionLib pages give a good overview on how-to get development started by providing workflows and tools. There is also a page collecting important news & updates of each release.
For troubleshooting, lookup the FAQ or write us a message at firstname.lastname@example.org. We’ve also made our documentation public available at docs.visionlib.com, so you don’t need to login repeatedly anymore.