Device databases provide Vuforia applications with a locally accessible database of image targets.
If you know what the application will be expected to see, and know that the target set of images is fewer than 1000, then a local device database is the correct solution. Device database use cases include promotional campaigns for a specific set of products or print material. An example is anything that will be identified and scanned with a lower set of images. Device databases receive more immediate responses because the images are stored locally.
See: Best practices for large device databases
Cloud databases provide Vuforia applications with a large number of targets. Cloud databases are stored in the Internet and support over a million image targets.
If the application supports a retail or catalog use case where there are more than 1000 images or the images are updated frequently, then a cloud database is the correct solution. Cloud image targets may take a little longer to identify, depending on network connectivity, but they provide a strong image recognition capability.
To help make a decision, consider the comparison of the two databases, as listed in the following table.
|Recommended for databases of 1000 targets or less
||Allows >1 million targets in the cloud database
|Allows downloading targets in different combinations
||Is made up of one database with all images and metadata
|Downloaded targets are only for detection, no metadata support
||Targets retrieved by cloud recognition can carry up to 1 MB of metadata
|Network connection on the end-user device is not required for detection
||Network connection on the end-user device is required for recognition, network traffic for cloud recognition
|Run time detection response time is within 2 to 3 frames
||Response time is affected by network conditions
|Multiple databases can be active (each with maximum 1000 targets)
||Recognition on maximum 1 million targets is active at any time
||Free and paid, depending on usage