Face redaction with deepstream

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Hi,
I’m running face redaction app on deepstream cloned from this NVIDIA-AI-IOT repo (GitHub - NVIDIA-AI-IOT/redaction_with_deepstream: An example of using DeepStream SDK for redaction). Is there a way to blur the face detected instead of just colouring it?

2

.

3

Hi,

Sorry, we only have the rectangle type example. You can just update the post process part on your own.
[url]redaction_with_deepstream/deepstream_redaction_app.c at master · NVIDIA-AI-IOT/redaction_with_deepstream · GitHub

Thanks.

4

osd_sink_pad_buffer_probe() function just passes the RGB values to a struct object NvOSD_RectParams which in turn passes the RGB values to a color struct NvOSD_ColorParams. My question is how and where are you actually using these values to draw a patch. Thanks

5

Hi,

Not sure if I understand your question correctly.

Suppose you can update the bbox color in the configure directly.
https://docs.nvidia.com/metropolis/deepstream/dev-guide/index.html#page/DeepStream_Development_Guide%2Fdeepstream_app_config.3.2.html%23wwpID0E0AB0HA

<b>bbox-bg-color</b>

The color of the boxes drawn over objects of a specific class ID, in RGBA format. The key 
must be of format bbox-bg-color<class-id>. This property can be used multiple times 
for multiple class IDs. If it is not used for a class ID, the boxes are not drawn for objects of 
that class ID.

If you just want some blur effect, you can set try to set the RGB into (255,255,255) and alpha to the value larger than 0.

Thanks.