Open anstosa opened 5 years ago
Would the MVP be just predicting "line present" vs. "line absent" for each camera view? I can help with that.
That would be amazing! My basic thinking is:
n
(TBD) sequential frames agree about the presence of a line (to drop false positives due to people driving the lane) then flag the line as up to that camera and automatically subtract that number of cars from the sailing space countsDoes that seem reasonable to you?
😨