I have successfully ran Peoplenet model on python example deepstream-test-1 . But there is weird behavior It show high number of objects than in the image. check the images below.
When I ran it on the original Peoplenet example using deepstream-app -c /opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/deepstream_app_source1_peoplenet.txt it works fine and number of objects seems same as in the image.
This is the config file I used
[property]
gpu-id=0
# preprocessing parameters:
net-scale-factor=0.0039215697906911373
batch-size=1
model-color-format=0
# Model specific paths. These need to be updated for every classification model.
tlt-model-key=tlt_encode
tlt-encoded-model=/opt/nvidia/deepstream/deepstream-5.0/samples/models/tlt_pretrained_models/peoplenet/resnet34_peoplenet_pruned.etlt
labelfile-path=/opt/nvidia/deepstream/deepstream-5.0/samples/models/tlt_pretrained_models/peoplenet/labels.txt
num-detected-classes=3
input-dims=3;544;960;0
uff-input-blob-name=input_1
output-blob-names=output_bbox/BiasAdd;output_cov/Sigmoid
force-implicit-batch-dim=1
## 0=FP32, 1=INT8, 2=FP16 mode
network-mode=2
# process-mode: 2 - inferences on crops from primary detector, 1 - inferences on whole frame
process-mode=1
interval=0
gie-unique-id=1
#scaling-filter=0
#scaling-compute-hw=0
[class-attrs-all]
pre-cluster-threshold=0.4
## Set eps=0.7 and minBoxes for cluster-mode=1(DBSCAN)
eps=0.7
minBoxes=1
What could be the problem? Is there any additional config parameters I need to add or change. I have doubt that class-attrs-all section needs to be changed but not sure.
Thanks in advance
Number of objects 30, Person_count=22, Face_count=8
Number of objects 32, Person_count=18, Face_count=14
