Secondary detector on the full frame

I try to deploy multiples detector models on Deepstream. e. g. Detector1–Detector2… I want all the detectors work on the full image frame instead of the upstream outputs. Where should I modify?

I am using TX2 with Deepstream 5. Here are the config.txt:
[application]
enable-perf-measurement=1
perf-measurement-interval-sec=5
#gie-kitti-output-dir=streamscl

[tiled-display]
enable=1
rows=1
columns=1
width=1280
height=720

[source0]
enable=1
#Type - 1=CameraV4L2 2=URI 3=MultiURI 4=RTSP 5=CSI
type=5
camera-csi-sensor-id=0
camera-width=1280
camera-height=720
camera-fps-n=30
camera-fps-d=1
camera-csi-sensor-id=0

[sink0]
enable=1
#Type - 1=FakeSink 2=EglSink 3=File 4=RTSPStreaming 5=Overlay
type=5
sync=1
display-id=0
offset-x=0
offset-y=0
width=0
height=0
overlay-id=1
source-id=0

[osd]
enable=1
border-width=2
text-size=15
text-color=1;1;1;1;
text-bg-color=0.3;0.3;0.3;1
font=Serif
show-clock=0
clock-x-offset=800
clock-y-offset=820
clock-text-size=12
clock-color=1;0;0;0

[streammux]
live-source=1
batch-size=6
batched-push-timeout=40000

Set muxer output width and height

width=1280
height=720

[primary-gie]
enable=1
batch-size=6
#Required by the app for OSD, not a plugin property
bbox-border-color0=1;0;0;1
bbox-border-color1=0;1;1;1
bbox-border-color2=0;0;1;1
bbox-border-color3=0;1;0;1
interval=0
#Required by the app for SGIE, when used along with config-file property
gie-unique-id=1
#config-file=pgie_ssd_tlt_config.txt
config-file= pgie.txt

[secondary-gie0]
enable=1
batch-size=6
gie-unique-id=2
#operate-on-gie-id=1
#operate-on-class-ids=0;
config-file= sgie.txt

[tests]
file-loop=0

here are the pgie.txt and sgie.txt:

pgie.txt
[property]
gpu-id=0
net-scale-factor=1.0
offsets=103.939;116.779;123.68
model-color-format=1
labelfile-path=./models/ssd/labels.txt
tlt-encoded-model=./models/ssd/ssd_resnet10_120.etlt
#model-engine-file=./models/ssd/ssd_resnet10_120.etlt_b8_gpu0_fp16_fire.engine

#tlt-model-key=nvidia_tlt
tlt-model-key=YOUR_KEY
uff-input-dims=3;544;960;0
uff-input-blob-name=Input
batch-size=1

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

network-mode=2
num-detected-classes=1
interval=0
gie-unique-id=1
is-classifier=0

#network-type=0
output-blob-names=NMS
parse-bbox-func-name=NvDsInferParseCustomSSDTLT
custom-lib-path=./nvdsinfer_customparser_ssd_tlt/libnvds_infercustomparser_ssd_tlt.so

[class-attrs-all]
pre-cluster-threshold=0.35
roi-top-offset=0
roi-bottom-offset=0
detected-min-w=0
detected-min-h=0
detected-max-w=0
detected-max-h=0

sgie.txt
[property]
gpu-id=0
net-scale-factor=1.0
offsets=103.939;116.779;123.68
model-color-format=1
labelfile-path=./models/ssd/labels.txt
tlt-encoded-model=./models/ssd/ssd_resnet10_120.etlt
model-engine-file=./models/ssd/ssd_resnet10_120.etlt_b6_gpu0_fp16.engine
tlt-model-key=YOUR_KEY
uff-input-dims=3;544;960;0
uff-input-blob-name=Input
batch-size=1

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

network-mode=2
num-detected-classes=1
interval=0
gie-unique-id=2

process-mode=1

#network-type=0
output-blob-names=NMS
parse-bbox-func-name=NvDsInferParseCustomSSDTLT
custom-lib-path=./nvdsinfer_customparser_ssd_tlt/libnvds_infercustomparser_ssd_tlt.so
pre-cluster-threshold=0.4

nvinfer work as full frame mode by default If you don’t set the process-mode in nvinfer config, what’s the issue you met?

The first detector works on the full frame. Now I want the second detector to work on the full frame, too. So should I just modify the process-mode value in nvinfer config?

Yes, pls refer nvinfer section https://docs.nvidia.com/metropolis/deepstream/plugin-manual/index.html#page/DeepStream%20Plugins%20Development%20Guide/deepstream_plugin_details.3.01.html#wwpID0E0OFB0HA