People count application With Deepstream SDK and Transfer Learning Toolkit

• Description
• Prerequisites
• Getting Started
• Build
• Run
• Output
• References

  

Description

This is a sample application for counting people entering/leaving in a building using NVIDIA Deepstream SDK, Transfer Learning Toolkit (TLT) and pre-trained models. This application can be used to build real-time occupancy analytics application for smart buildings, hospitals, retail, etc. The application is based on deepstream-test5 sample application.

It takes streaming video as input, counts the number of people crossing a tripwire and sends the live data to the cloud. In this application, you will learn:

• How to use PeopleNet model from NGC
• How to use NvDsAnalytics plugin to draw line and count people crossing the line

• How to send the analytics data to cloud or another microservice over Kafka

  You can extend this application to change region of interest, use cloud-to-edge messaging to trigger record in the DeepStream application or build analytic dashboard or database to store the metadata.

To learn how to build this demo step-by-step, check out the on-demand webinar on Creating Intelligent places using DeepStream SDK.

Prerequisites

• Install Deepstream: [https://docs.nvidia.com/metropolis/deepstream/dev-guide/index.html#page/DeepStream_Development_Guide/deepstream_quick_start.html#]

• Download PeopleNet model: [https://catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/models/peoplenet/files]

• This application is based on deepstream-test5 application. More about test5 application: [https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_ref_app_test5.html]

• Install Kafka: [https://kafka.apache.org/quickstart] and create the kafka topic:

  tar -xzf kafka_2.13-3.5.0.tgz

  cd kafka_2.13-3.5.0

  bin/zookeeper-server-start.sh config/zookeeper.properties

  bin/kafka-server-start.sh config/server.properties

  bin/kafka-topics.sh --create --topic quickstart-events --bootstrap-server localhost:9092

Getting Started

• Preferably clone the repo in $DS_SDK_ROOT/sources/apps/sample_apps/
• Download peoplnet model: cd deepstream-occupancy-analytics/config && ./model.sh
• For Jetson use: bin/jetson/libnvds_msgconv.so
• For x86 use: bin/x86/libnvds_msgconv.so

Build and Configure

• Set CUDA_VER in the MakeFile as per platform.

  For Jetson, CUDA_VER=11.4

  For x86, CUDA_VER=11.8

  cd deepstream-occupancy-analytics && make

• Set msg-conv-msg2p-lib at [sink1] group in dstest_occupancy_analytics.txt as per platform

  For Jetson

  msg-conv-msg2p-lib=$DEEPSTREAM_SDK_PATH/deepstream-occupancy-analytics/bin/jetson/libnvds_msgconv.so

  For x86

  msg-conv-msg2p-lib=$DEEPSTREAM_SDK_PATH/deepstream-occupancy-analytics/bin/x86/libnvds_msgconv.so

Run

./deepstream-test5-analytics -c config/dstest_occupancy_analytics.txt

In another terminal run this command to see the kafka messages:

bin/kafka-console-consumer.sh --topic quickstart-events --from-beginning --bootstrap-server localhost:9092

Output

The output will look like this:

Where you can see the kafka messages for entry and exit count.

References

• CREATE INTELLIGENT PLACES USING NVIDIA PRE-TRAINED VISION MODELS AND DEEPSTREAM SDK: [https://info.nvidia.com/iva-occupancy-webinar-reg-page.html?ondemandrgt=yes]
• Deepstream SDK: [https://developer.nvidia.com/deepstream-sdk]
• Deepstream Quick Start Guide: [https://docs.nvidia.com/metropolis/deepstream/dev-guide/index.html#page/DeepStream_Development_Guide/deepstream_quick_start.html#]
• Transfer Learning Toolkit: [https://developer.nvidia.com/transfer-learning-toolkit]