Deepstream6.0.1 python api How do I choose to present the results of my reasoning

[tang_gouwu]

Please provide complete information as applicable to your setup.

**• Hardware Platform： Jetson
**• DeepStream Version：6.0.1
• JetPack Version (valid for Jetson only)
**• TensorRT Version：8.2

Deepstream6.0.1 how to modify the python API obj_label information, or what is selective in the output shown in part the result of the reasoning results, for example: reasoning out someone and car, I just want to only show people in the output screen as a result, the need to do.

Car doesn’t want to show, it just wants to show Person

[yuweiw]

The simpler method is to use a model that only detects people.
If you want to delete the other object in the image, you need to traverse all the display meta and delete the other objects except people.

[tang_gouwu]

The model cannot be changed, so how do you iterate and remove other elements using the deepstream python api???

[yuweiw]

Could you try to set the filter-out-class-ids first?
For example, if the class-id of person in your demo is 0. You can set that to filter-out-class-ids=0;.

[tang_gouwu]

Thanks! The problem has been solved. What do I do if I want to change the value of obj_label, for example, from Person to man。。

[yuweiw]

You can change the words in you label.txt file.

[tang_gouwu]

Neither the model nor label.txt can be changed, so it can only be changed through post-processing

[tang_gouwu]

Neither the model nor label.txt can be changed, so it can only be changed through post-processing。

[yuweiw]

Why can’t you change the label file? This is just a text file, which is used to configure your label.

[tang_gouwu]

Just tried to modify, thank you for your support