Facial Landmarks NVIDIA

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Hello
I am working on AGX and I am trying to make my model work with Facial Landmarks Estimation from ngc with deepstream.
My primary consists of the facenet model provided by ngc and I am trying to make the secondary with the following configurations:

[property]
gpu-id=0
net-scale-factor= 0.0039215686274
#force-implicit-batch-dim=1
tlt-model-key=nvidia_tlt
tlt-encoded-model=/model.etlt
uff-input-blob-name=input_1
batch-size=1

0=FP32 and 1=INT8 mode

network-mode=2
#infer-dims=3;40;160
#input-object-min-width=30
#input-object-min-height=30
input-dims=1;80;80;0
output-blob-names=output_bbox/BiasAdd;output_cov/Sigmoid
process-mode=2
model-color-format=1
gpu-id=0
gie-unique-id=2
operate-on-gie-id=1
operate-on-class-ids=0
is-classifier=1
classifier-async-mode=0
classifier-threshold=0.51
process-mode=2
output-tensor-meta=1
#scaling-filter=1
#scaling-compute-hw=0

However it is failing to build the engine. Are there any others parameters to focus on?

3

Hi,

Could you share your environment with us first?

• Hardware Platform (Jetson / GPU)
• DeepStream Version
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs)
• 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)

More, could you share what kind of TensorRT error do you meet?
Thanks.

4

Hardware Platform: Jetson nano AGX
DeepStream Version: 5.0
Jetpack version: 4.4
TensorRT: 7.1.3

I would like to know what are the configurations of the secondary model for the facial landmarks since it is failing to build the model.

5

Hi,

Could you also share the error message with us?
Since the cause of failure may vary from the use case.

Thanks.

6

ERROR: [TRT]: UffParser: Could not read buffer.
parseModel: Failed to parse UFF model
ERROR: failed to build network since parsing model errors.
ERROR: Failed to create network using custom network creation function
ERROR: Failed to get cuda engine from custom library API
0:00:01.818591012 17494 0x8385150 ERROR nvinfer gstnvinfer.cpp:613:gst_nvinfer_logger: NvDsInferContext[UID 2]: Error in NvDsInferContextImpl::buildModel() <nvdsinfer_context_impl.cpp:1735> [UID = 2]: build engine file failed
Segmentation fault (core dumped)

7

Hi,

Could you share the complete log with us?

It seems that TensorRT meet some issue when open the given uff file.
Do you update the path accordingly?

Thanks.

8

it works now, I followed the steps mentioned in
https://docs.nvidia.com/metropolis/TLT/tlt-user-guide/text/tlt_cv_inf_pipeline/quick_start_scripts.html#tlt-cv-quick-start-scripts
and created the engine.
Thank you!

10

Hi @melissa1 ,
I am facing similar issue as your while working on this can you please share your config file if possible that you used to setup facial landmark model.

Thanks