FAQ

We are very focused on solutions that allow users to seamlessly see customized content behind any object or logo they capture. We believe that QR codes go against such seamless experience and therefore we do not support them.

The Image Recognition part and the Augmented Reality part have different requirements.

The Image Recognition component works with any object, regardless of its real-world geometry, by design. The technology is engineered to recognize not only planar objects such as a book covers, but also rounded objects such as cans and bottles. You can read more

We have focused our development efforts on recognizing objects with a specific pattern or texture. Face recognition is a mature technology that you can surely find off-the-shelf in the market.

Currently the objects are being recognized based mainly on grayscale information, also known as the intensity or luminance component.

​In many use-cases using all color components would not help. In some cases, in uncontrolled lighting conditions full color may harm the recognition. Also, using luminance helps us keep the response times low. But w

Currently we do not perform Optical Character Recognition (OCR) of the scene text.

​Moreover, all large paragraphs of dense text, without any image or logo, look similar to our system and should be avoided. On the other hand large titles above images or other graphical elements are visually very distinctive and usually help in the page recognition

Yes, but you need to add two versions of the reference image, the original one and the flipped one, to the item that you wish to recognize.

All query images are being saved. Currently you cannot access them via our APIs, but you can contact us to get them.

Currently you must upload the actual image file, irrespectively whether you use the Management API or the web panel. However, if you already have your own tools for downloading image files from urls you can extend them with our Python or PHP scripts for image file upload.

Most types of objects can be identified with our Image and Object Recognition technology.
Image and Object recognition can be performed on objects such as:

CD/DVD and book covers
Newspapers and magazines
Logos and branding
Posters and Packaged goods
Monuments and landmarks

Objects which have a repetitive pattern like a wooden table or

Our SDKs include a synchronization feature to keep on-device collection bundles up to date.
“Sync” compares the collection on the device with the one stored in CraftAR and downloads the items that have been changed.
This method works on all our SDKs. However, at the moment it does not download the augmented reality contents. We recommend its use fo

YES. Both CraftAR service and the SDKs, (Android & iOS AR and Pro SDKs, and Cordova SDK) support videos edited in a special way to include a transparency mask. To know more about how we support videos with transparency in the CraftAR Service and in the SDKs, see how to add videos with transparency to Augmented Reality with CraftAR .

You can download the SDKs of your choice for free from within CraftAR. For being able to use those SDKs in published apps, please contact our team to get more information about our plans.

Yes, as our service allows image recognition applied to your own database. In order to create collections and add items to those collections, you can use the CraftAR web panel or the Management API.

Our service performs the recognition within less than 500 ms.

The total recognition round-trip depends on your network bandwidth. Normally on Wi-Fi and 4G you get the results within 600-800 ms. On 3G the round-trip varies from 1 to 2 seconds.

A Token is a string of characters that uniquely identifies the collection of images that you want to search through. Tokens are used in the Image Recognition API, in the CraftAR Mobile App, and in the CraftAR mobile SDKs to specify that collection.

If you’re trying to find the default token for your collection, follow these steps:

Log in to the

Tags allow you to group items inside a specific collection. Then, you can create on-device bundles including one of those subsets. (You can download several on-device bundles to a device, but you can only use one tag for each bundle.)

You can assign tags to specific items based on the logic that your app requires. For example, you can differentiat