I'm working on an AI that can detect people and certain vehicles from a camera feed. The user can configure which of these "objects" they want to monitor. They can choose either people, vehicles or both.
When enabling a category, a "drawer" is opened. See picture below. There I want to allow the user to:
- In the people category, specify which wearables to look for or ignore.
- In the vehicles category, specify which types of vehicles to look for or ignore.
- Per default, both categories are set to detect everything, without any of these "filters".
I started with the design of the vehicles drawer, which currently seems to solve the use-case pretty well: The user can specify which types of vehicles to look for and which ones to ignore.
However, things became a bit trickier when I started designing the drawer for the people category, because unlike vehicles which only belong to exactly one sub-type (e.g a vehicle cannot be a car and a bus at the same time), people can obviously wear wear none, 1, 2 or all 3 wearables at the same time. Also, if you have all 3 enabled, does it mean that someone needs to wear all of them or any one?
I also want the user to be able to specify if the selected attributes are the ones the AI should look for or if it is the ones it should ignore.
This is important because some construction sites require people to wear certain gear when entering the site, while others require people not to wear certain gear when entering (e.g ear protection can be dangerous when workers need to be attentive to their surroundings).