These sorts of systems tend to be far easier to trick than humans with certain sorts of "camouflage". Is the reverse true? Are there some kinds of markings that confuse people but not modern image recognition systems?
The thing that concerns me more is the machine's confidence. A human seeing unusual lane markings would be confused and exercise extra caution. Confusing a person is easy, but I don't know how you could make them…
It doesn't really cost money to move data around. The biggest cost by far is building the infrastructure. If the capacity is available, the cost of using it is minimal.
These sorts of systems tend to be far easier to trick than humans with certain sorts of "camouflage". Is the reverse true? Are there some kinds of markings that confuse people but not modern image recognition systems?
The thing that concerns me more is the machine's confidence. A human seeing unusual lane markings would be confused and exercise extra caution. Confusing a person is easy, but I don't know how you could make them…
It doesn't really cost money to move data around. The biggest cost by far is building the infrastructure. If the capacity is available, the cost of using it is minimal.