This video contains 90 frames. For some reason, the DeepStream test 3 sample app and other apps that iterate through NvDsBatchMeta.frame_meta_list
processes 1 frame more than the actual. See DeepStream test 3 sample app result below
Frame Number = 0 Number of objects = 6 Vehicle Count = 0 Person Count = 6
Frame Number = 1 Number of objects = 4 Vehicle Count = 0 Person Count = 4
Frame Number = 2 Number of objects = 5 Vehicle Count = 0 Person Count = 5
Frame Number = 3 Number of objects = 6 Vehicle Count = 0 Person Count = 6
Frame Number = 4 Number of objects = 5 Vehicle Count = 0 Person Count = 5
Frame Number = 5 Number of objects = 5 Vehicle Count = 0 Person Count = 5
...
Frame Number = 86 Number of objects = 6 Vehicle Count = 0 Person Count = 6
Frame Number = 87 Number of objects = 6 Vehicle Count = 0 Person Count = 6
Frame Number = 88 Number of objects = 5 Vehicle Count = 0 Person Count = 5
Frame Number = 89 Number of objects = 4 Vehicle Count = 0 Person Count = 4
Frame Number = 90 Number of objects = 4 Vehicle Count = 0 Person Count = 4
it should have been Frame 0 - Frame 89
. What might be the reason of this?
Environment
Architecture: x86_64
GPU: NVIDIA GeForce GTX 1650 Ti with Max-Q Design
NVIDIA GPU Driver: Driver Version: 495.29.05
DeepStream Version: 6.0 (running on docker image nvcr.io/nvidia/deepstream:6.0-devel)
TensorRT Version: v8001
Issue Type: Question