Kakip Board

Prerequisites

Ensure the following:

• Complete the Getting Started instructions provided by Renesas.
• The Kakip board is set up, including the preparation of the SD card.
• The rzv2h_ai_sdk_image Docker container is running on the host machine.

Kakip Board Resources

For additional information on the Kakip board, please refer to:

• Website: Kakip Board Official Website
• GitHub Repository: Kakip Board GitHub Repository

Refer to the image below for a visual guide to setting up the Kakip board for development:

Head Count Top View Application

This application is used to count the human heads present in a video from Image/camera input.

This repository contains the 11_Head_count_topview application. Below are the detailed steps for setting up, applying patches, running the application, and building it.

Step 1: Refer to the SDK Getting Started Guide

Visit the Renesas RZ/V AI SDK Getting Started Guide for the setup.

Step 2: Clone the Repository

It is recommended to download/clone the repository on the data folder which is mounted on the rzv2h_ai_sdk_container docker container as shown below.

   cd <path_to_data_folder_on_host>/data
   git clone https://github.com/Ignitarium-Renesas/rzv_ai_apps.git
   git clone https://github.com/Ignitarium-Renesas/Kakip_RZV2H_Demos.git

Note 1: Please verify the git repository url if error occurs.

Note 2: This command will download whole repository, which include all other applications.
If you have already downloaded the repository of the same version, you may not need to run this command.

Step 3: Start the Docker Container

Run (or start) the docker container and open the bash terminal on the container.
Here, we use the rzv2h_ai_sdk_container as the name of container, created from rzv2h_ai_sdk_image docker image.
> Note that all the build steps/commands listed below are executed on the docker container bash terminal.

Step 4: Set Environment Variables

Set your clone directory to the environment variable.

    export PROJECT_PATH=/drp-ai_tvm/data

Step 5: Apply the Patch File

Apply the provided patch file to the application using the command below:

cd ${PROJECT_PATH}/rzv_ai_apps/11_Head_count_topview/
git apply ../../Kakip_RZV2H_Demos/Head_count_topview/Head_count_topview.patch

Step 6: Navigate to the Application Directory

Move to the source code directory of the application:

    cd ${PROJECT_PATH}/rzv_ai_apps/11_Head_count_topview/src

Step 7: Build the Application

Build the application by following the commands below.

    mkdir -p build && cd build
    cmake -DCMAKE_TOOLCHAIN_FILE=./toolchain/runtime.cmake -DV2H=ON ..
    make -j$(nproc)

Step 8: Locate the Generated Application

The built application file will be available in the following directory:

   ${PROJECT_PATH}/11_Head_count_topview/src/build

The generated file will be named:

    head_count_topview_app

Copy the Object File (optional)

Running the Program Using the Precompiled Object File

If you prefer to skip the compilation process, you can directly use the precompiled object file included in this repository. This step is useful if you want to quickly run the program without setting up the build environment.

cp -r ../../Kakip_RZV2H_Demos/Head_count_topview/head_count_topview_app exe_v2h

Application: Deploy Stage

For the ease of deployment all the deployables file and folders are provided on the exe_v2h folder.

File	Details
topview_head_count_yolov3	Model object files for deployment.
head_count_topview_app	application file.

1. Follow the steps below to deploy the project on the board.
   1. Run the commands below to download the 11_Head_count_topview_deploy_tvm-v230.so from Release v5.00

   cd ${PROJECT_PATH}/11_Head_count_topview/exe_v2h/topview_head_count_yolov3
   wget https://github.com/Ignitarium-Renesas/rzv_ai_apps/releases/download/v5.00/11_Head_count_topview_deploy_tvm-v230.so
   

   2. Rename the 11_Head_count_topview_deploy_tvm-v230.so to deploy.so.

   mv 11_Head_count_topview_deploy_tvm-v230.so deploy.so
   

   3. Copy the following files to the /home/root/tvm directory of the rootfs (SD Card) for the board.
      • All files in exe_v2h directory. (Including deploy.so file.)
      • 11_Head_count_topview application file if you generated the file by following all the steps.

Run the Application

To run the application on the Kakip board:

On the board terminal, execute the application with the following command, specifying either USB camera or IMAGE input mode:

• Image Input:

  ./head_count_topview_app IMAGE <path_to_the_image>

• USB Camera Input:

  ./head_count_topview_app USB

Animal detection Application

This application detects animals using a YOLOv3 model and classifies them based on input from a USB camera or an image file.

Steps Overview

Follow Steps 1–4 from the Head Count Top View Guide:

• Set up the environment, clone repositories, and configure the Docker container.
• Ensure that the PROJECT_PATH is set:

  export PROJECT_PATH=/drp-ai_tvm/data

Step 5: Apply the Patch File

Apply the provided patch file to the application using the command below:

cd ${PROJECT_PATH}/rzv_ai_apps/07_Animal_detection/
git apply ../../Kakip_RZV2H_Demos/Animal_detection/animal_detection.patch

Step 6: Navigate to the Application Directory

Move to the source code directory of the application:

    cd ${PROJECT_PATH}/rzv_ai_apps/07_Animal_detection/src

Follow Step 7 from the Head Count Top View Guide to build the application

Step 8: Locate the Generated Application

The built application file will be available in the following directory:

   ${PROJECT_PATH}/07_Animal_detection/src/build

The generated file will be named:

    animal_detection_app

Copy the Object File (optional)

Running the Program Using the Precompiled Object File

If you prefer to skip the compilation process, you can directly use the precompiled object file included in this repository. This step is useful if you want to quickly run the program without setting up the build environment.

cp -r ../../Kakip_RZV2H_Demos/Animal_detection/animal_detection_app exe_v2h

Application: Deploy Stage

For the ease of deployment all the deployables file and folders are provided on the exe_v2h folder.

File	Details
animal_yolov3_onnx	Model object files for deployment.
animal_detection_app	application file.

1. Follow the steps below to deploy the project on the board.
   1. Run the commands below to download the 07_Animal_detection_deploy_tvm-v230.so from Release v5.00

   cd ${PROJECT_PATH}/07_Animal_detection/exe_v2h/animal_yolov3_onnx
   wget https://github.com/Ignitarium-Renesas/rzv_ai_apps/releases/download/v5.00/07_Animal_detection_deploy_tvm-v230.so
   

   2. Rename the 07_Animal_detection_deploy_tvm-v230.so to deploy.so.

   mv 07_Animal_detection_deploy_tvm-v230.so deploy.so
   

   3. Copy the following files to the /home/root/tvm directory of the rootfs (SD Card) for the board.
      • All files in exe_v2h directory. (Including deploy.so file.)
      • 07_Animal_detection application file if you generated the file by following all the steps.

Run the Application

To run the application on the Kakip board:

On the board terminal, execute the application with the following command, specifying either USB camera or IMAGE input mode:

• Image Input:

  ./animal_detection_app IMAGE <path_to_the_image>

• USB Camera Input:

  ./animal_detection_app USB

Multi camera vehicle detection Application

This application is used to detect 10 types of vehicles below from camera input.
Also it can be used for these vehicles at 360 angle with multi cameras.

• Car, policecar, ambulance, bicycle, bus, truck, bike, tractor , auto and fire engine

Steps Overview

Follow Steps 1–4 from the Head Count Top View Guide:

• Set up the environment, clone repositories, and configure the Docker container.
• Ensure that the PROJECT_PATH is set:

  export PROJECT_PATH=/drp-ai_tvm/data

Step 5: Apply the Patch File

Apply the provided patch file to the application using the command below:

cd ${PROJECT_PATH}/rzv_ai_apps/14_Multi_camera_vehicle_detection/
git apply ../../Kakip_RZV2H_Demos/multi_camera_vehicle_detection/multi_camera.patch

Step 6: Navigate to the Application Directory

Move to the source code directory of the application:

    cd ${PROJECT_PATH}/rzv_ai_apps/14_Multi_camera_vehicle_detection/src

Follow Step 7 from the Head Count Top View Guide to build the application

Step 8: Locate the Generated Application

The built application file will be available in the following directory:

   ${PROJECT_PATH}/14_Multi_camera_vehicle_detection/src/build

The generated file will be named:

    multi_camera_vehicle_detection_app

Copy the Object File (optional)

Running the Program Using the Precompiled Object File

If you prefer to skip the compilation process, you can directly use the precompiled object file included in this repository. This step is useful if you want to quickly run the program without setting up the build environment.

cp -r ../../Kakip_RZV2H_Demos/multi_camera_vehicle_detection/multi_camera_vehicle_detection_app exe_v2h

Application: Deploy Stage

For the ease of deployment all the deployables file and folders are provided on the exe_v2h folder.

File	Details
Multi_camera_vehicle_detection_tinyyolov3	Model object files for deployment.
multi_camera_vehicle_detection_app	application file.

1. Follow the steps below to deploy the project on the board.

   1. Verify the presence of deploy.so file in ${PROJECT_PATH}/14_Multi_camera_vehicle_detection/exe_v2h/Multi_camera_vehicle_detection_tinyyolov3

   2. Copy the following files to the /home/root/tvm directory of the rootfs (SD Card) for the board.

      • All files in exe_v2h directory. (Including deploy.so file.)
      • 14_Multi_camera_vehicle_detection application file if you generated the file by following all the steps.

Run the Application

To run the application on the Kakip board:

On the board terminal, execute the application with the following command, specifying the number cameras as the 2nd argument and use 'FLIP' as the third argument if you want to flip the output.

• Application with USB Camera Input:

  ./multi_camera_vehicle_detection_app USB 2

• Application with USB Camera Input with flip mode:

  ./multi_camera_vehicle_detection_app USB 2 FLIP

Hand gesture recognition Application

This application is not covered with MIT license. This application is licensed with Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) license. Please have a look at dos and dont's here : Creative commons website link Hand gesture model's reference : Dataset link

User can : Share — copy and redistribute the material in any medium or format Adapt — remix, transform, and build upon the material for any purpose, even commercially.

Under these terms: Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use. ShareAlike — If you remix, transform, or build upon the material, you must distribute your contributions under the same license as the original. No additional restrictions — You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.

This application showcases the capability of deep neural networks to predict different hand gestures. It detect total of 8 Gestures that includes one, two, three, four, five, thumbs up, thumbs down and rock in the hand with the highest precision.

Steps Overview

Follow Steps 1–4 from the Head Count Top View Guide:

• Set up the environment, clone repositories, and configure the Docker container.
• Ensure that the PROJECT_PATH is set:

  export PROJECT_PATH=/drp-ai_tvm/data

Step 5: Apply the Patch File

Apply the provided patch file to the application using the command below:

cd ${PROJECT_PATH}/rzv_ai_apps/12_Hand_gesture_recognition_v2/
git apply ../../Kakip_RZV2H_Demos/Hand_gesture/hand_gesture.patch

Step 6: Navigate to the Application Directory

Move to the source code directory of the application:

    cd ${PROJECT_PATH}/rzv_ai_apps/12_Hand_gesture_recognition_v2/src

Follow Step 7 from the Head Count Top View Guide to build the application

Step 8: Locate the Generated Application

The built application file will be available in the following directory:

   ${PROJECT_PATH}/12_Hand_gesture_recognition_v2/src/build

The generated file will be named:

    hand_gesture_recognition_v2_app

Copy the Object File (optional)

Running the Program Using the Precompiled Object File

If you prefer to skip the compilation process, you can directly use the precompiled object file included in this repository. This step is useful if you want to quickly run the program without setting up the build environment.

cp -r ../../Kakip_RZV2H_Demos/Hand_gesture/hand_gesture_recognition_v2_app exe_v2h

Application: Deploy Stage

For the ease of deployment all the deployables file and folders are provided on the exe_v2h folder.

File	Details
hand_yolov3_onnx	Model object files for deployment.
hand_gesture_recognition_v2_app	application file.

1. Follow the steps below to deploy the project on the board.

   1. Run the commands below to download the 12_Hand_gesture_recognition_v2_deploy_tvm-v230.so from Release v5.00

   cd ${PROJECT_PATH}/12_Hand_gesture_recognition_v2/exe_v2h/hand_yolov3_onnx
   wget https://github.com/Ignitarium-Renesas/rzv_ai_apps/releases/download/v5.00/12_Hand_gesture_recognition_v2_deploy_tvm-v230.so
   

   2. Rename the 12_Hand_gesture_recognition_v2_deploy_tvm-v230.so to deploy.so.

   mv 12_Hand_gesture_recognition_v2_deploy_tvm-v230.so deploy.so
   

   3. Verify the presence of deploy.so file in ${PROJECT_PATH}/12_Hand_gesture_recognition_v2/exe_v2h/hand_yolov3_onnx
   4. Copy the following files to the /home/root/tvm directory of the rootfs (SD Card) for the board.
      • All files in exe_v2h directory. (Including deploy.so file.)
      • 12_Hand_gesture_recognition_v2 application file if you generated the file by following all the steps.

Run the Application

To run the application on the Kakip board:

On the board terminal, execute the application with the following command, specifying either USB camera or IMAGE input mode:

• Image Input:

  ./hand_gesture_recognition_v2_app IMAGE <path_to_the_image> 

• USB Camera Input:

  ./hand_gesture_recognition_v2_app USB