OpenMMLab Video Perception Toolbox. It supports Video Object Detection (VID), Multiple Object Tracking (MOT), Single Object Tracking (SOT), Video Instance Segmentation (VIS) with a unified framework.
I have installed mmtrack and want to use it to count people crossing a virtual line (entering a building, for example). However, when I read all the documentation, I could not find any method that is able to interact with specific object coardinates, which is required to count line crossings. Then I could hardly find any way to interact with the information I get, now it seems to me that your library is just a way to show how models work. But it's not, is it? Can you tell me more about it, maybe you can give me some more code examples?
I have installed mmtrack and want to use it to count people crossing a virtual line (entering a building, for example). However, when I read all the documentation, I could not find any method that is able to interact with specific object coardinates, which is required to count line crossings. Then I could hardly find any way to interact with the information I get, now it seems to me that your library is just a way to show how models work. But it's not, is it? Can you tell me more about it, maybe you can give me some more code examples?