How to push the stream of several camera to cloud or local singly

Please provide complete information as applicable to your setup

• Hardware Platform (Jetson / GPU)
x86-64 Ubuntu 20.04 LTS machine with Geforce GTX 3060
• DeepStream Version
6.1
• TensorRT Version
8.2.5.1
• NVIDIA GPU Driver Version (valid for GPU only)
515.48.07
• Issue Type( questions, new requirements, bugs)
I have several camera, and I want to push the stream of them to the local or cloud singly, the example seems only able to push one stream or all of them together, is there a way to push the stream of several cameras to the local or cloud ,which I can use to handle the stream singly

what does that mean about “is there a way to push the stream of several cameras to the local or cloud”? deepstream can process one stream or many streams, please refer to sample deepstream-test3.

deepstream has an example to show how to push inference dstream using sink component, but it only push stream of one or streams of all cameras together to cloud, but I want to get the separate inference stream to push to the cloud.

which example? do you mean pushing inference results to cloud server?

Yes,I mean how to push inference results to cloud server, for example source30_1080p_dec_infer-resnet_tiled_display_int8.txt, enable sink2 and choose type=4 RTSPStreaming

it only push stream of one or streams of all cameras together to cloud server, but I want to get the separate inference stream to push to the cloud server.

please refer to How to configure different Kafka topics for different sources on deepstream, you can set source-id in [sink] configuration.

what if I only want to push inference results to localhost, for example rtsp://localhost:8554/ds-test, and every camera has an unique rtsp address, what should I do?

you can configure multiple sinks, in every sink, set type=4, and set different rtsp-port and udp-port.

[sink0]
enable=1
#Type - 1=FakeSink 2=EglSink 3=File 7=nv3dsink (Jetson only)
type=2
sync=1
source-id=0
gpu-id=0
nvbuf-memory-type=0

[sink1]
enable=1
type=4
#1=mp4 2=mkv
container=1
#1=h264 2=h265
codec=1
#encoder type 0=Hardware 1=Software
enc-type=0
sync=0
#iframeinterval=10
bitrate=4000
#H264 Profile - 0=Baseline 2=Main 4=High
#H265 Profile - 0=Main 1=Main10
profile=0
output-file=out.mp4
source-id=0

set below properties in case of RTSPStreaming

rtsp-port=8554
udp-port=5400

[sink2]
enable=1
#Type - 1=FakeSink 2=EglSink 3=File 4=RTSPStreaming
type=4
#1=h264 2=h265
codec=1
#encoder type 0=Hardware 1=Software
enc-type=0
source-id=1
sync=0
bitrate=4000
#H264 Profile - 0=Baseline 2=Main 4=High
#H265 Profile - 0=Main 1=Main10
profile=0

set below properties in case of RTSPStreaming

rtsp-port=8555
udp-port=5401

I have several camera, and I want to get camera 0 inference result by set source-id =0 of sink1, get camera 1 inference result by set source-id =1 of sink2,but it did not work, how can I get different inference result through different rtsp-port?

please provide the whole configuration file.

After I set enable=0 in tiled-display group, it works,and I can get camera1 inference result by setting source-id=1 in sink group. if I set enable=1 in tiled-display group, is the whole tiled-display image seems as source-id 0,and there is no source-id 1?

test5 supports enable_with_demux + link-to-demux, it will output tiler and one source, please refer to test5_config_file_src_infer_tiler_demux.txt