search

openremote / or-objectdetection

object detection in video
Other
94 stars 51 forks source link

How to change resolution of output video and the tracking Line #2

Open Tensorhunt opened 4 years ago

Tensorhunt commented 4 years ago

How to change resolution of output video and the tracking Line?

Can you please add in which files do we update it?

mesakarghm commented 4 years ago

What type of video are you trying on? I am trying it on prerecorded videos and I am not getting any detections at all? The code seems alright to me.

richturner commented 4 years ago

FYI: This was a proof of concept project carried out by a group of students and therefor support isn't possible (unless one of the students involved are able to help).

sieuwe1 commented 4 years ago

@Tensorhunt

Changing the resolution is difficult since the line is hardcoded and the yolo model output is transformed to a certain resolution. You can change this but then you have to find something online. I recommend you leave the opencv windows the same resolution as they are now. And just resize you video image at the beginning to be the video resolution we used. I think at the end we used a 4k camera. But if that resolution does not work try 1080p.