Realtek AmebaPro2 NN deployment

• NN Deployment
  • Using customized NN model
    • Setup Acuity toolkit on PC
    • Step for customized model conversion
    • Supported model quantization type on NN accelerator
  • SDK configuration for customized NN model
    • NN related file in SDK
    • Add customized model network binary to SDK
    • Create a model object can be used by VIPNN module
  • NN memory and flash usage evaluation
    • Evaluate memory usage of model
    • Evaluate model size on flash
  • Using the NN MMF example with VIPNN module
    • Set RGB video resolution as model input size
    • Choose NN model
    • Build NN example
    • Update NN model on flash
    • Validate NN example
  • How to add a pre-process node to customized model (optional)
  • How to load model from SD card instead of flash (optional)
  • How to modify customized model name after conversion (optional)
• Model Security
  • Model Authentication – Hash and Signature check
  • Model Encryption – Cypher Text Decryption
  • SDK Configuration
  • Secure Deployment Flow
  • Signing and Encryption PC Tool
  • Performance Test
• Post-process PC Development Tool
• Appendix A. Acuity Supported Operation Layer
  • ONNX to ACUITY Operation Mapping
  • Darknet to ACUITY Operation Mapping