I has error when run with tensorRT 8.5.2.2 or any 8.5.x
phongans opened this issue · comments
root@phong:~/VideoPipe/build/bin# ./multi_trt_infer_nodes_sample
[2024-04-30 13:38:44.278][Info ][7f2c6f2c5000][/root/VideoPipe/nodes/vp_image_src_node.cpp:34] [image_src_0] [multifilesrc location=./vp_data/test_images/vehicle/%d.jpg loop=true ! jpegparse ! jpegdec ! videorate ! video/x-raw,framerate=1/1 ! videoconvert ! appsink]
[2024-04-30 13:38:44.279][Warn ][7f2c6f2c5000][/root/VideoPipe/nodes/vp_infer_node.cpp:43] [trt_detector] cv::dnn::readNet load network failed!
[2024-04-30 13:38:45.028][Warn ][7f2c6f2c5000][/root/VideoPipe/nodes/vp_infer_node.cpp:43] [trt_color_cls] cv::dnn::readNet load network failed!
[2024-04-30 13:38:45.084][Warn ][7f2c6f2c5000][/root/VideoPipe/nodes/vp_infer_node.cpp:43] [trt_type_cls] cv::dnn::readNet load network failed!
[2024-04-30 13:38:45.144][Warn ][7f2c6f2c5000][/root/VideoPipe/nodes/vp_infer_node.cpp:43] [trt_encoder] cv::dnn::readNet load network failed!
[2024-04-30 13:38:45.256][Info ][7f2c6f2c5000][/root/VideoPipe/nodes/vp_screen_des_node.cpp:14] [screen_des_0] [appsrc ! videoconvert ! videoscale ! textoverlay text=screen_des_0 halignment=left valignment=top font-desc='Sans,16' shaded-background=true ! timeoverlay halignment=right valignment=top font-desc='Sans,16' shaded-background=true ! queue ! fpsdisplaysink video-sink=ximagesink sync=false]
[ WARN:0@1.055] global /home/phong/Projects/opencv-4.6.0/modules/videoio/src/cap_gstreamer.cpp (1405) open OpenCV | GStreamer warning: Cannot query video position: status=0, value=-1, duration=-1
terminate called after throwing an instance of 'cv::Exception'
what(): OpenCV(4.6.0) /home/phong/Projects/opencv-4.6.0/modules/imgproc/src/resize.cpp:4055: error: (-215:Assertion failed) inv_scale_x > 0 in function 'resize'
Aborted (core dumped)
seems it was not caused by tensorrt, it crashed becausecv:: resize failed.
which test video do you test? video file in vp_data from google drive? also make sure vp_data at the same directory where you run samples.
Hi,
i tried trt model but it seems it is not compatible with cuda version.
Thanks
Get Outlook for Android<https://aka.ms/AAb9ysg>
…________________________________
From: zhouzhi ***@***.***>
Sent: Sunday, May 5, 2024 8:59:04 PM
To: sherlockchou86/VideoPipe ***@***.***>
Cc: Phong Pham ***@***.***>; Author ***@***.***>
Subject: Re: [sherlockchou86/VideoPipe] I has error when run with tensorRT 8.5.2.2 or any 8.5.x (Issue #43)
seems it was not caused by tensorrt, it crashed becausecv:: resize failed.
which test video do you test? video file in vp_data from google drive?
—
Reply to this email directly, view it on GitHub<#43 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/A5ZOFL7FW27K5YMWI6EOAZDZAY3KRAVCNFSM6AAAAABG7SDQUSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOJUHAYTSOJQGU>.
You are receiving this because you authored the thread.Message ID: ***@***.***>
yes, you'd better run samples based on opencv::dnn which uses onnx models.
How can I run your trt model for license plate recognition testing?