We are using the Milestone VPS plugin with Deepstream Pipeline that includes different Detection/Classification Models.
When showing the bounding boxes on the live preview of the camera, the bounding boxes of the detection are shown perfectly. However, when going to the playback preview, the bounding boxes are no longer aligned to the object that it is detected upon.
Please note that we are using Milestone 2020 R2 and VPS From MIPSDK 2020 R1. Milestone is running on a Windows OS while the VPS/Deepstream Pipeline is running on Ubuntu 18.04.
Are these informative or do you require more info about the setup?
Please note that the issue was caused by difference in the clock between the server of deepstream and the server of Milestone. Using an NTP server to synchronize them solved the issue .