Face_Reid_Surveillance_System

Near Real-Time Face-reIdentification Surveillance System Pipeline

Install latest version of docker

• Install Docker - if Docker not installed on your machine yet
• Ensure your linux user is in the group docker

Install JAVA 11

• Install openjdk 11

Download vitis-ai docker

docker pull xilinx/vitis-ai:1.4.1.978 

Clone this repository

git clone --recurse-submodules https://github.com/durgabhavaniv/Face_Reid_Surveillance_System.git

cd Face_Reid_Surveillance_System

VCK5000 setup

• Download below files from, https://www.xilinx.com/member/vck5000.html#vitis, and follow below commads
  • xrt_202020.2.9.317_20.04-amd64-xrt.deb
  • xilinx-vck5000-es1-gen3x16-platform-2-1_all.deb.tar.gz
  • xilinx-vck5000-es1-gen3x16-2-202020-1-dev_1-3123623_all.deb

sudo apt-get insall ./xrt_202020.2.9.317_20.04-amd64-xrt.deb

tar -xzvf xilinx-vck5000-es1-gen3x16-platform-2-1_all.deb.tar.gz

cd ./xilinx-vck5000-es1-gen3x16-platform-2-1_all.deb/

sudo apt-get install ./xilinx-*

cd ..

sudo apt-get install ./xilinx-vck5000-es1-gen3x16-2-202020-1-dev_1-3123623_all.deb

cd ~/Face_Reid_Surveillance_System/setup/vck5000

source ./install.sh

Run docker for VCK5000

cd Face_Reid_Surveillance_System

./docker_run.sh xilinx/vitis-ai:1.4.1.978

Update the stored faces data

cd ~/Face_Reid_Surveillance_System/demo/Vitis-AI-Library/stored_faces/

Follow below setup commands inside docker

sudo apt-get update

sudo apt-get install libcanberra-gtk-module -y

source /workspace/setup/vck5000/setup.sh

sudo mkdir /usr/share/vitis_ai_library/models

cd /workspace/demo/Vitis-AI-Library/samples/reid

sudo cp reid /usr/share/vitis_ai_library/models -r

cd /workspace/demo/Vitis-AI-Library/samples/facedetect

sudo cp densebox_640_360 /usr/share/vitis_ai_library/models -r

Rebuilding face reidentification in your system incase of cpp binaries are not supported.

cd /workspace/demo/Vitis-AI-Library/samples/reid

bash -x build.sh

cd /workspace/demo/Vitis-AI-Library/samples/facedetect

sudo cp ./demo.hpp /usr/include/vitis/ai/demo.hpp

bash -x build.sh

Run person reidentification demo via CLI using following commads

./run.sh /workspace/demo/Vitis-AI-Library/input/Gate.mp4 &
./run.sh /workspace/demo/Vitis-AI-Library/input/Hall_entry.mp4 &
./run.sh /workspace/demo/Vitis-AI-Library/input/Room_entry.mp4 &
./run.sh /workspace/demo/Vitis-AI-Library/input/Room_exit.mp4 &
./run.sh /workspace/demo/Vitis-AI-Library/input/Hall_exit.mp4

Results via CLI

• ~/Face_Reid_Surveillance_System/demo/Vitis-AI-Library/output this directory givens the deetcted faces results.
• ~/Face_Reid_Surveillance_System/demo/Vitis-AI-Library/output/output.txt gives results of tracking of person faces with respect to camera number and time.

Run person reidentification demo with frontend application and neo4j

• Edit 'Face_Reid_Surveillance_System/blob/master/hackster/src/main/resources/application.properties' file, by modifing directory path based on your system. (Ex: \home\durga\Face_Reid_Surveillance_System to \your_directory_path\Face_Reid_Surveillance_System)

./start_application.sh

Page1 : video upload and video connections

Page2 : input videos validation

Page3 : search queries for processed output

System requirements

• os version - Ubuntu 20.04 LTS
• kernel version - 5.4.0-52-generic

Complete description of this project

• Use below link https://www.hackster.io/honeygo/near-real-time-face-reidentification-surveillance-system-d16eaf#toc-let-s-get-start-with-face-reidentification-surveillance-system-10