Filters in "deepstream-infer-tensor-meta-test" don't do anything

[aidan3]

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) GPU
• DeepStream Version 6.1
• JetPack Version (valid for Jetson only)
• TensorRT Version 8+
• NVIDIA GPU Driver Version (valid for GPU only) 510.85
• Issue Type( questions, new requirements, bugs) Question
• 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)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)

Hi,

I am testing out the “deepstream_infer_tensor_meta.cpp” example as I am looking to implement something similar, and noticed for both the pgie, and sgies the “class-attrs-all” and “classifier-threshold” respectively don’t actually do anything. Is there any way to get them to actually be applied? It would just be nicer then adding them in manually.

FYI, to determine that they weren’t applied I…

1. Set “detectionParams.perClassPreclusterThreshold = {0.0, 0.0, 0.0, 0.0};” on line 210 in “deepstream_infer_tensor_meta_test.cpp”

2. Set attributes in “[class-attrs-all]” in “dstensor_pgie_config.txt” as

[class-attrs-all]
pre-cluster-threshold=0.99999
topk=1
nms-iou-threshold=0.01

3.) Set “classifier-threshold=0.99” in all sgie configs.

[fanzh]

network-type=100
output-tensor-meta=1
if network-type is 100, you need to implement post processing in callback function, in pgie_pad_buffer_probe, sgie_pad_buffer_probe, you can define own perClassPreclusterThreshold .

[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