[PR] Implement post-process of 3d pose estimation (SimpleBaseline)

[tucan9389]

demo

Related Issue

• Support SimpleBaselinePoseEsitmation tflite model (3D-pose) #28

PR Points

• Implement the pre-process
  • Image normalization using in PyTorch usually
• Implement the post-process
  • Define the keypoints each index
  • Implement softmax
  • Implement sum function that could get dim as augment like numpy or pytorch
  • Optimize sum and mul with Accelerate framework
• Log the 3d points
• Draw the 3d points in the app screen
• Support real-time (later)
• Implement additional post-processing for removing some noises (later)

[tucan9389]

Currently, the output of the post-processing is incorrect now. I'm asking the post-processing method to the model engineer now.

[tucan9389]

I implemented the 3d pose estimation post-processing now. I fixed the original argmax to softargmax.