File size: 2,753 Bytes
a3aa9ee b33b869 a3aa9ee b33b869 a3aa9ee |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 |
---
license: other
license_link: >-
https://github.st.com/AIS/stm32ai-modelzoo/raw/master/pose_estimation/LICENSE.md
---
# Head landmarks quantized
## **Use case** : `Pose estimation`
# Model description
Head landmarks is a single pose estimation model targeted for real-time processing implemented in ONNX.
The model is quantized in int8 format using onnx quantizer.
## Network information
| Network information | Value |
|-------------------------|-----------------|
| Framework | ONNX Runtime |
| Quantization | int8 |
| Provenance | https://github.com/PINTO0309/PINTO_model_zoo/tree/main/032_FaceMesh
| Paper | https://developers.google.com/ml-kit/vision/face-mesh-detection |
## Networks inputs / outputs
With an image resolution of NxM with K keypoints to detect :
| Input Shape | Description |
| ----- | ----------- |
| (1, N, M, 3) | Single NxM RGB image with UINT8 values between 0 and 255 |
| Output Shape | Description |
| ----- | ----------- |
| (1, 1, 1, Kx2) | FLOAT values Where Kx2 are the (x,y) values of each keypoints |
## Recommended Platforms
| Platform | Supported | Recommended |
|----------|-----------|-------------|
| STM32L0 | [] | [] |
| STM32L4 | [] | [] |
| STM32U5 | [] | [] |
| STM32H7 | [] | [] |
| STM32MP1 | [x] | [] |
| STM32MP2 | [x] | [x] |
| STM32N6 | [x] | [x] |
# Performances
## Metrics
Measures are done with default STM32Cube.AI configuration with enabled input / output allocated option.
### Reference **NPU** memory footprint
|Model | Format | Resolution | Series | Internal RAM (KiB) | External RAM (KiB) | Weights Flash (KiB) | STEdgeAI Core version |
|----------|--------|-------------|------------------|------------------|---------------------|-------|-------------------------|
| [head_landmarks](https://github.com/STMicroelectronics/stm32ai-modelzoo/tree/main/pose_estimation/headlandmarks/custom_dataset_heads_468kpts/facelandmarksv1_192_int8.onnx) | Int8 | 224x224x3 | STM32N6 | 1739.5 | 0.0 | 3246.47 | 3.0.0 |
### Reference **NPU** inference time
| Model | Format | Resolution | Board | Execution Engine | Inference time (ms) | Inf / sec | STEdgeAI Core version |
|--------|--------|-------------|------------------|------------------|---------------------|-------|-------------------------|
| [head_landmarks](https://github.com/STMicroelectronics/stm32ai-modelzoo/tree/main/pose_estimation/headlandmarks/custom_dataset_heads_468kpts/facelandmarksv1_192_int8.onnx) | Int8 | 224x224x3 | STM32N6570-DK | NPU/MCU | 20.52 | 48.73 | 3.0.0 |
|