Emotion net config for Deepstream

Please provide the following information when requesting support.

• Hardware (T4/V100/Xavier/Nano/etc)
• Network Type (Detectnet_v2/Faster_rcnn/Yolo_v4/LPRnet/Mask_rcnn/Classification/etc)
• TLT Version (Please run “tlt info --verbose” and share “docker_tag” here)
• Training spec file(If have, please share here)
• How to reproduce the issue ? (This is for errors. Please share the command line and the detailed log here.)

I am trying to run Emotion net on deepstream:

enginePath=…/…/…/…/models/emotion/emotion.etlt_b32_gpu0_fp16.engine
etltPath=…/…/…/…/models/emotion/emotion.etlt
etltKey=nvidia_tlt

networkMode can be int8, fp16 or fp32

networkMode=fp16
batchSize=32
inputWidth=136
inputHeight=1
inputType=
label=0,neutral
label=1,happy
label=2,surprise
label=3,squint
label=4,disgust
label=5,scream

Is this config correct??

OR is this the config file??

[property]
gpu-id=0
net-scale-factor=0.0039215697906911373
#model-file=/opt/nvidia/deepstream/deepstream-6.1/deepstream_tao_apps/apps/tao_detection/Age-and-Gender-Recognition/models/age_net.caffemodel
#proto-file=/opt/nvidia/deepstream/deepstream-6.1/deepstream_tao_apps/apps/tao_detection/Age-and-Gender-Recognition/models/age_deploy.prototxt
model-engine-file=/opt/nvidia/deepstream/deepstream-6.1/model.etlt_b32_gpu0_int8.engine
tlt-model-key=nvidia_tlt
tlt-encoded-model=/opt/nvidia/deepstream/deepstream-6.1/model.etlt
labelfile-path=/opt/nvidia/deepstream/deepstream-6.1/labels_emotion.txt
#int8-calib-file=/opt/nvidia/deepstream/deepstream-6.1/deepstream_tao_apps/apps/tao_detection/int8_calibration.txt
#enginePath=…/…/…/…/models/emotion/emotion.etlt_b32_gpu0_fp16.engine
#etltPath=…/…/…/…/models/emotion/emotion.etlt
etltKey=nvidia_tlt

#dynamic batch size
batch-size=32

0=FP32, 1=INT8, 2=FP16 mode

network-mode=1
num-detected-classes=1
#output-blob-names=softargmax,softargmax:1,conv_keypoints_m80
output-blob-names = input_landmarks:0,softmax/Softmax:0
#0=Detection 1=Classifier 2=Segmentation 100=other
network-type=100

Enable tensor metadata output

output-tensor-meta=1
#1-Primary 2-Secondary
process-mode=2
gie-unique-id=2
operate-on-gie-id=1
net-scale-factor=1.0
offsets=0.0
input-object-min-width=5
input-object-min-height=5
#0=RGB 1=BGR 2=GRAY
model-color-format=2

[class-attrs-all]
threshold=0.0

What should I add, and what should I delete?
On the deepstream tao_apps it is said for emotion net, we can use Facial landmarks config, but when I use that there is a binding error, what is the actual config file contents???

Refer to deepstream_tao_apps/apps/tao_others at release/tao3.0_ds6.0.1 · NVIDIA-AI-IOT/deepstream_tao_apps · GitHub

A sample of emotions:

./deepstream-emotion-app 1 ../../../configs/facial_tao/sample_faciallandmarks_config.txt file:///usr/data/faciallandmarks_test.jpg ./emotion

This is what I am trying to tell…
The config file is from Facial landmarks!!!
The contents of the faciallandmarks.txt file::::
numLandmarks=80
maxBatchSize=32
inputLayerWidth=80
inputLayerHeight=80

This is of no use to me.

What is this config then?

enginePath: …/…/…/…/models/emotion/emotion.etlt_b32_gpu0_fp16.engine
etltPath: …/…/…/…/models/emotion/emotion.etlt
etltKey: nvidia_tlt

networkMode can be int8, fp16 or fp32

networkMode: fp16
batchSize: 32
inputWidth: 136
inputHeight: 1
inputType:
label:

  • 0,neutral
  • 1,happy
  • 2,surprise
  • 3,squint
  • 4,disgust
  • 5,scream

Please refer to the master branch.

A sample of emotions:

./deepstream-emotion-app 1 ../../../configs/facial_tao/sample_faciallandmarks_config.txt file:///usr/data/faciallandmarks_test.jpg ./emotion

or ./deepstream-emotion-app emotion_app_config.yml

Please check if you can run with the default setting successfully.

Have you seen the config file for the same?

I am trying to Run a Primary model [Face], then Run a Secondary Classifier [Emotion net]…
The secondary is failing!!

I have been looking into this repo, and what I don’t understand is why use faciallandmark config for emotion net.

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

Please refer to deepstream_tao_apps/apps/tao_others/deepstream-emotion-app at master · NVIDIA-AI-IOT/deepstream_tao_apps · GitHub
The application pipeline graph

emotion application pipeline

Please check if you can run with the default setting successfully.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.