Pre-trained Free AI Application Libraries for RZ/V2L

General Information

• Free, open-source based library of pre-trained AI applications available on github.
• ⭐Source Code: https://github.com/Ignitarium-Renesas/RZV2L_AiLibrary
• This Library has API functions for leveraging AI applications that will run on Renesas RZ/V2L Board. Currently this library has following sample applications:
  • Human Head Counter
  • Line crossing object Counter
  • Elderly people fall detection (Work in progress)
  • Safety helmet and vest detection
  • Human age and gender detection (Work in progress)
  • Face recognition and spoof detection (Work in progress

Pre-Trained AI Article

Addition Notes

The Pre-trained models include pre-compiled applications as well as AI Models translated to run on the DRP-AI hardware. These files are located in the "exe" folder for each Pre-trained Application. These precompiled application are compiled for the Renesas RZV2L EVK using the Coral MIPI Camera. This folder can be simple copied to the board using SCP recursive command. NOTE : Some Pre-trained Applicatino

Support USB Camera

By default the Pretrained Applications are compiled to use the MIPI camera. The Pre-trained applications can be modified to use USB camera. This modification is only relevant to applications that support video.

1. Open the application folder src folder. For example 01_Head_count/Head_count_cam/src
2. Open the define.h header file.
3. Find the following line. Comment out the macro that defines INPUT_CORAL.

/* Coral Camera support */
#define INPUT_CORAL

Sample Videos

NOTE: Memory Usage includes the Image Input, Inference Output, Inference Weights and Inference Parameters. Applications that use multiple AI Models are run sequentially

Head Count Application Model : YoloV3 Memory Usage: 235MB Inference Input Shape : 416,416,3 Inference Time: 348ms	Head Count
Line Crossing Object Counting Model : TinyYoloV2 Memory Usage: 52MB Inference Input Shape : 416,416,3 Inference Time:	Line Count
Fall Detection Model : Tiny Yolov2 Memory Usage: 52MB Inference Input Shape : 256, 192, 3 Inference Time: 59ms Model : HRNET Memory Usage: 129MB Inference Input Shape : 256, 192, 3 Inference Time: 163ms Total Inference Time : 222ms	Fall Detection
Age and Gender Detection Model : Custom age Memory Usage: 24MB Inference Input Shape : 416,416,3 Inference Time: 10ms Model : Custom gender Memory Usage: 87MB Inference Input Shape : 224,224,3 Inference Time: 10ms Model : Tiny YoloV2 Memory Usage: 52MB Inference Input Shape : 416,416,3 Inference Time: 59ms Total Inference Time: 79ms	Age and Gender Detection
Face Recognition, Spoofing, and Registration Model : Resnet50 Memory Usage: 88MB Inference Input Shape : 224,224,3 Inference Time: 96ms	Face Recognition
Animal Detection Model : YoloV3 Memory Usage: 236MB Inference Input Shape : 416,416,3 Inference Time: 360ms	Animal Detection
Hand Gesture Recognition Model : Custom Pose Detector for Hand Memory Usage: 91MB Inference Input Shape : 256,256,3 Inference Time: 256ms	Hand Gesture Recognition
Human Gaze Recognition Model : Resnet18 Memory Usage: 38MB Inference Input Shape : 416,416,3 Inference Time: 33 Model : Tiny YoloV2 Memory Usage: 52MB Inference Input Shape : 416,416,3 Inference Time: 58 Total Inference Time: 91ms	Gaze Recognition