How to replace the buffer data?

[yangyi]

Please provide complete information as applicable to your setup.

• Hardware Platform (GPU)
**• DeepStream Version6.1
**• TensorRT Version*8.2
**• NVIDIA GPU Driver Version (valid for GPU only)11.6-515

Dear professor:
Thank you for your reading this information. I hope you have a good day.
I meet a problem shown in the following picture.

I cropped the same area in 100 frames of the stream, and sum, average to get the up picture. So I hope to send the picture to the secondary engine.
All the operations are in dsexample. and the pipeline is the default setting.

I hope to use the picture to replace the original buffer data, and send the location information to the secondary engine.
Could you give me some advice?
(1) Is the method right? If it is right, how to replace the buffer data? could you gave me some reference?
(2) If the method is hard to realize, what I should do?

Thank you very much.

[fanzh]

1. what is your use case? what is your media pipeline?
2. please refer to dsexample plugin, it can blur the object area, you can add a object meta, and fill the specific data.

[yangyi]

Thank you for your response.
(1) My case is material measurement on the belt conveyor. The belt moves quickly, so I need to stack the belt area more than 100 frames, and get the average. The stack result is sent to secondary engine to measure the material.
(2) My pipeline is:
Cameras → StreamMux → PremayGIE → tracking → dsexample → secondaryGIE → sink
(3) Thank you advise the blur function. Can the blur result be used to infer in secondary engine? I will try .

Thank you very much

[fanzh]

did you still meet any deepstream issue?

[fanzh]

There is no update from you for a period, assuming this is not an issue any more.
Hence we are closing this topic. If need further support, please open a new one.
Thanks