DLI Nano class: Getting Started with DeepStream for Video Analytics on Jetson Nano

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Nano B01, DLI Class " Getting Started with DeepStream for Video Analytics on Jetson Nano

I just made SD card from the course.

2.0 Multiple Networks Application for make secondary inference engine (automaker) crashed as soon as I execute using the Jupyter notebook.

However, color and type SGIEs are together working fine.

This, I copied the solution coming with the notebook. Then I compiled and executed. It also provides the same error. Thus, I suspect that the automaker SGIE seems having a bug.

Please verify my observation.
If so, is there any way to go around or fix this?

Here is the runtime error of the auto maker SGIE case:

/home/dlinano/deepstream_sdk_v4.0.2_jetson/sources/apps/my_apps/dst2-two-sgie
With tracker

*** DeepStream: Launched RTSP Streaming at rtsp://localhost:8554/ds-test ***

Now playing: /home/dlinano/deepstream_sdk_v4.0.2_jetson/samples/streams/sample_720p.h264
Opening in BLOCKING MODE
Opening in BLOCKING MODE
Creating LL OSD context new
0:00:01.021934924 13795 0x5582b85330 WARN nvinfer gstnvinfer.cpp:515:gst_nvinfer_logger: NvDsInferContext[UID 3]:useEngineFile(): Failed to read from model engine file
0:00:01.022005290 13795 0x5582b85330 INFO nvinfer gstnvinfer.cpp:519:gst_nvinfer_logger: NvDsInferContext[UID 3]:initialize(): Trying to create engine from model files
0:00:01.022178678 13795 0x5582b85330 WARN nvinfer gstnvinfer.cpp:515:gst_nvinfer_logger: NvDsInferContext[UID 3]:generateTRTModel(): INT8 not supported by platform. Trying FP16 mode.
0:00:01.022235190 13795 0x5582b85330 ERROR nvinfer gstnvinfer.cpp:511:gst_nvinfer_logger: NvDsInferContext[UID 3]:generateTRTModel(): Cannot access caffemodel file ‘/home/dlinano/deepstream_sdk_v4.0.2_jetson/sources/apps/my_apps/dst2-two-sgie/…/…/…/…/samples/models/Secondary_CarMaker/resnet18.caffemodel’
0:00:01.022265763 13795 0x5582b85330 ERROR nvinfer gstnvinfer.cpp:511:gst_nvinfer_logger: NvDsInferContext[UID 3]:initialize(): Failed to create engine from model files
0:00:01.022308837 13795 0x5582b85330 WARN nvinfer gstnvinfer.cpp:692:gst_nvinfer_start: error: Failed to create NvDsInferContext instance
0:00:01.022333160 13795 0x5582b85330 WARN nvinfer gstnvinfer.cpp:692:gst_nvinfer_start: error: Config file path: dstest2_sgie2_config.txt, NvDsInfer Error: NVDSINFER_CONFIG_FAILED
Running…
ERROR from element secondary2-nvinference-engine: Failed to create NvDsInferContext instance
Error details: /dvs/git/dirty/git-master_linux/deepstream/sdk/src/gst-plugins/gst-nvinfer/gstnvinfer.cpp(692): gst_nvinfer_start (): /GstPipeline:dstest2-pipeline/GstNvInfer:secondary2-nvinference-engine:
Config file path: dstest2_sgie2_config.txt, NvDsInfer Error: NVDSINFER_CONFIG_FAILED
Returned, stopping playback
Deleting pipeline

2

I have carefully looked at the color, automaker, vehicletype run.
I thought that it worked, but it was not. The color infor is used twice and vehicletype is appended at the end. Definitely, automaker SGIE does not work Here is the screen shot. It shows: color, color, and vehicle type.

Please help me how to fix. I think it is a bug of automaker SGIE.

image
image968×474 120 KB

4

Could you share your setup info, seems you are using DS 4, we suggest you to use DS 5.0.
For your question:

  1. firstly you need to make sure the sgie model output is correct
  2. Can you try deepstream-test2 to see if that work