4 ms·
In this example you could query for "poop" and then look at the resulting plot for a sustained period of time with high poop probability? :)
by axg11 5y ago
In this example you could query for "poop" and then look at the resulting plot for a sustained period of time with high poop probability? :)
- grepfru_it 5y agohow do you distinguish between poop and mud? :) its great in theory but real world creates harder challenges. the real answer is to keep track of all the dogs that pass by and then compare the before/after scenes for differences. i'm doing this with my garage door camera for detecting water (for a completely different problem)
- axg11 5y agoPerhaps analyze for the combination of dogs + poop? The dog probability will be transient/a spike, whereas the poop probability will be constant (after appearance of the dog).