I am having trouble getting some image targets to work and be detected. How can I fix this?
For images to work as a target, they need to have a minimum level of detail in order for the system to be detected. A plain image such as a white wall would not have enough detail. Also black, clear or reflective images and objects make bad targets as the imaging system will often be dazzled by these.
The best images should be flat and have many details to be picked up properly by the recognition system. The best picture is not always the most detailed picture, experimenting with different resolutions can often get better target matching.
The target needs to occupy at least one third of the camera to be detected properly.
I am having trouble getting an image that was being recognized before to be recognized again. How can I fix this?
There can be several reasons for this issue. This can be caused by having duplicate experiences targeting the same image. It also can be caused by dramatic changes in lighting or the object not being flat enough. Diverium image recognition tries to find an “exact” match to the image that it originally saw. While there are features to compensate for different angles and lighting conditions, anything that dramatically changes the appearance of the target may cause it to be undetected.