Facial-Landmark-Detection Example

This document describes how to use the QAI AppBuilder Python API to run inference with the Facial-Landmark-Detection facial landmark detection model on Qualcomm® Hexagon™ Processor (NPU).

Supported Devices

Device	SoC
Fogwise® AIRbox Q900	QCS9075

Install QAI AppBuilder

tip

1. Please install QAI AppBuilder according to QAI AppBuilder Installation Guide.

2. Please configure ADSP environment variables according to Create ADSP Environment Variables.

Run the Example

Install Dependencies

Device

pip3 install requests tqdm qai-hub py3_wget opencv-python torch torchvision matplotlib easyocr qai_hub_models

Run the Script

• Navigate to the example directory

  QCS9075

  Device

  cd ai-engine-direct-helper/samples/python

• Prepare input image. The following image is used as an example:

  

  input image

• Run inference

  Device

  python3 facemap_3dmm/facemap_3dmm.py

  $ python3 facemap_3dmm/facemap_3dmm.py
       0.0ms [WARNING]  <W> Initializing HtpProvider
  
  /prj/qct/webtech_scratch20/mlg_user_admin/qaisw_source_repo/rel/qairt-2.37.1/point_release/SNPE_SRC/avante-tools/prebuilt/dsp/hexagon-sdk-5.5.5/ipc/fastrpc/rpcmem/src/rpcmem_android.c:38:dummy call to rpcmem_init, rpcmem APIs will be used from libxdsprpc
       0.0ms [WARNING]  <W> This META does not have Alloc2 Support
  
       0.0ms [WARNING]  <W> This META does not have Alloc2 Support
  
       0.0ms [WARNING]  <W> This META does not have Alloc2 Support
  
       0.0ms [WARNING]  <W> This META does not have Alloc2 Support
  
     120.2ms [WARNING] Time: Read model file to memory. 9.63
  
       0.0ms [WARNING]  <W> This META does not have Alloc2 Support
  
       0.0ms [WARNING]  <W> This META does not have Alloc2 Support
  
       0.0ms [WARNING]  <W> This META does not have Alloc2 Support
  
       0.0ms [WARNING]  <W> This META does not have Alloc2 Support
  
       0.0ms [WARNING]  <W> This META does not have Alloc2 Support
  
       0.0ms [WARNING]  <W> This META does not have Alloc2 Support
  
       0.0ms [WARNING]  <W> This META does not have Alloc2 Support
  
     138.7ms [WARNING] Time: contextCreateFromBinary. 18.41
  
     138.8ms [WARNING] Time: UnmapViewOfFile. 0.00
  
     140.3ms [WARNING] Time: model_initialize facemap_3dmm 140.23
  
     182.0ms [WARNING] Time: model_inference facemap_3dmm 1.43
  
  /mnt/ssd/qualcomm/702/zzf_fork/test/ai-engine-direct-helper/samples/python/facemap_3dmm/facemap_3dmm.py:145: UserWarning: Creating a tensor from a list of numpy.ndarrays is extremely slow. Please consider converting the list to a single numpy.ndarray with numpy.array() before converting to a tensor. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:253.)
    _output = torch.tensor(output)
  Saving image to /mnt/ssd/qualcomm/702/zzf_fork/test/ai-engine-direct-helper/samples/python/facemap_3dmm/output.jpg
       0.0ms [WARNING]  <W> This META does not have Alloc2 Support
  
       0.0ms [WARNING]  <W> This META does not have Alloc2 Support
  
       0.0ms [WARNING]  <W> This META does not have Alloc2 Support
  
       0.0ms [WARNING]  <W> This META does not have Alloc2 Support
  
  /usr/bin/xdg-open: 882: www-browser: not found
  /usr/bin/xdg-open: 882: links2: not found
  /prj/qct/webtech_scratch20/mlg_user_admin/qaisw_source_repo/rel/qairt-2.37.1/point_release/SNPE_SRC/avante-tools/prebuilt/dsp/hexagon-sdk-5.5.5/ipc/fastrpc/rpcmem/src/rpcmem_android.c:42:dummy call to rpcmem_deinit, rpcmem APIs will be used from libxdsprpc
     706.9ms [WARNING] Time: model_destroy facemap_3dmm 72.10
  
  /usr/bin/xdg-open: 882: elinks: not found
  /usr/bin/xdg-open: 882: links: not found
  /usr/bin/xdg-open: 882: lynx: not found
  /usr/bin/xdg-open: 882: w3m: not found
  xdg-open: no method available for opening '/tmp/tmpkpcumu3y.PNG'

  

  output image