How do I replace data in a pipeline in DeepStream?

[1017948396]

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU)
RTX3060
• DeepStream Version
DeepStream Version: 6.2
• JetPack Version (valid for Jetson only)
• TensorRT Version
TensorRT Version 8.5
• NVIDIA GPU Driver Version (valid for GPU only)
525.85.12

The output frame of my detection network is not rendered directly on the image, but on a black image with a size of 200×400, how can I replace the image data in the deepstream pipeline.

[fanzh]

1. what is the media pipeline? what are the model’s inputs and outputs? here is a sample to modify image data :Deepstream sample code snippet - #3 by bcao

@1017948396

but on a black image with a size of 200×400

2. why is it a black image? the detection models usually give a bonding boxes, which are some coordinates.

[1017948396]

my model infer image.

[fanzh]

here is another solution, you might get the coordinates of that green bbox, convert them to new coordinates in the frame, add object meta by nvds_add_obj_meta_to_frame, then deepstream nvosd plugin will draw the bbox.

[1017948396]

He was painting the bbox on top of the original image without that black image.

[fanzh]

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

1. what are the model 's input and output layers?
2. about the black picture above, How did it come? how did that green bboxs come?