search

aws-samples / amazon-rekognition-video-analyzer

A working prototype for capturing frames off of a live MJPEG video stream, identifying objects in near real-time using deep learning, and triggering actions based on an objects watch list.
Other
368 stars 158 forks source link

Duplicate SNS text #42

Open dickiesanders opened 6 years ago

dickiesanders commented 6 years ago

I have this demo code up and running against my front door IP camera. It's working well, but when a person is in more than one frame, I get multiple SMS notifications. What's the best way to de-dup notifications?

Would it be possible to group frames and send one message?

knkinfy commented 5 years ago

Hi... Did you manage to find the solution ?

sumanghosh12 commented 4 years ago

@dickiesanders - Looks like this code is working fine for you. Did you checkin your code in GIT. Just want to refer it. I had to make some minute changes as I am using Pythin3.7 and this code is for Python2.7.

Web UI is up but empty in the dotted area.

When I run the pynt videocapture[10], I can see response coming back with correct labes.

But I am not able to see any frames in my frame-bucket.

Any help will be helpful.