openjetson / tensorrt-yolov5 Goto Github PK

in yolov5.cpp, line 526:
'''
auto start = std::chrono::system_clock::now();
doInference(*context, stream, buffers, data, prob, BATCH_SIZE);
auto end = std::chrono::system_clock::now();
int fps = 1000.0/std::chrono::duration_caststd::chrono::milliseconds(end - start).count();
'''
end-start is only the infer time, time for pre and post process should be included when calculating the FPS，like below:
'''
auto start1 = std::chrono::system_clock::now();
for (int f = 0; f < (int)file_names.size(); f++) {
... ...
doInference(*context, stream, buffers, data, prob, BATCH_SIZE);
auto end1 = std::chrono::system_clock::now();
int fps = 1000.0/std::chrono::duration_caststd::chrono::milliseconds(end1 - start1).count();
start1 = std::chrono::system_clock::now();
}
in this way end-start contains time for one loop.

Hi @tuyaliang, I tried using this script using my USB webcam and it worked perfectly, however when trying it using my Raspberry Pi v2 CSI camera, I ran into the following error and a pop-up window opened displaying a green screen:

skymind@skymind:~/yolov5-tensorrt-openjetson/build$ sudo ./yolov5 -v
[ WARN:0] global /home/nvidia/host/build_opencv/nv_opencv/modules/videoio/src/cap_gstreamer.cpp (1757) handleMessage OpenCV | GStreamer warning: Embedded video playback halted; module v4l2src0 reported: Internal data stream error.
[ WARN:0] global /home/nvidia/host/build_opencv/nv_opencv/modules/videoio/src/cap_gstreamer.cpp (886) open OpenCV | GStreamer warning: unable to start pipeline
[ WARN:0] global /home/nvidia/host/build_opencv/nv_opencv/modules/videoio/src/cap_gstreamer.cpp (480) isPipelinePlaying OpenCV | GStreamer warning: GStreamer: pipeline have not been created
Gtk-Message: 16:18:00.782: Failed to load module "canberra-gtk-module"

Would really appreciate any help in solving this problem. Thanks.