Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU) Orin NX / Orin NX • DeepStream Version 6.2 / 6.3 • JetPack Version (valid for Jetson only) 5.1 / 5.1.2 • TensorRT Version 8.5.2.2 / 8.5.2.2 • NVIDIA GPU Driver Version (valid for GPU only) Orin NX / Orin NX • Issue Type( questions, new requirements, bugs) bugs • How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
Using nvds_obj_enc_process() to save jpeg files may lead to the situation shown in the picture below.
In my testing, the occurrence probability of this phenomenon is approximately 0.07%.
And it only occurs on machines running Deepstream 6.3 / Jetpack 5.1.2.
Additionally, the following is the pipeline I used in the deepstream-app.
From your pipeline, where did you use the nvds_obj_enc_process API? And how did you confirm that it was an issue with this API, not with the stream itself?
I use the nvds_obj_enc_processAPI in the gie_primary_processing_done_buf_prob() function.
And sorry for any confusion. When using nvds_obj_enc_processAPI, I encountered those images. I’m not sure if there is an issue with the API. It is possible that, as you mentioned, the issue could be with the stream itself. How can I verify this?
You can try to add a tee plugin after the multi-src-bin plugin to create a new branch. And use the new branch to save the video source. Then compare the video source and the jpg image.
There are no examples for your use case. But we have many examples about how to use tee plugin. You can refer to our sources\apps\sample_apps\deepstream-test4\deepstream_test4_app.c source code.
There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks
There is a similar topic 279814.
Could you reproduce that with our demo code sources\apps\sample_apps\deepstream-image-meta-test?
Since our latest version DeepStream 6.4 has been released, could you reproduce that on the latest version too?