This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision | ||
|
products:sbc:edge2:npu:demos:retinaface [2023/08/25 06:23] louis |
products:sbc:edge2:npu:demos:retinaface [2025/04/09 23:33] (current) louis |
||
|---|---|---|---|
| Line 1: | Line 1: | ||
| - | ====== Demo5 - Retinaface ====== | + | ~~tag> NPU RetinaFace Edge2 PyTorch~~ |
| - | ===== Get Source Code ===== | + | ====== RetinaFace PyTorch Edge2 Demo - 5 ====== |
| - | The codes we use. | + | {{indexmenu_n> |
| + | |||
| + | ===== Introduction ===== | ||
| + | |||
| + | RetinaFace is a face detection model. It can draw five key points on each face, including two eyes, nose and two corners of mouth. | ||
| + | |||
| + | Inference results on Edge2. | ||
| + | |||
| + | {{: | ||
| + | |||
| + | **Inference speed test**: USB camera about **39ms** per frame. MIPI camera about **33ms** per frame. | ||
| + | |||
| + | ===== Train Model ===== | ||
| + | |||
| + | The codes we use [[gh> | ||
| ```shell | ```shell | ||
| git clone https:// | git clone https:// | ||
| + | ``` | ||
| + | |||
| + | Before training, modify '' | ||
| + | |||
| + | ```diff | ||
| + | diff --git a/ | ||
| + | index 87bb528..4a22f2a 100644 | ||
| + | --- a/ | ||
| + | +++ b/ | ||
| + | @@ -25,5 +25,6 @@ def get_lr(optimizer): | ||
| + | | ||
| + | |||
| + | def preprocess_input(image): | ||
| + | - image -= np.array((104, | ||
| + | + image = image / 255.0 | ||
| + | | ||
| ``` | ``` | ||
| Line 52: | Line 82: | ||
| ==== Convert ==== | ==== Convert ==== | ||
| - | |||
| - | Before training, modify '' | ||
| - | |||
| - | ```diff | ||
| - | diff --git a/ | ||
| - | index 87bb528..4a22f2a 100644 | ||
| - | --- a/ | ||
| - | +++ b/ | ||
| - | @@ -25,5 +25,6 @@ def get_lr(optimizer): | ||
| - | | ||
| - | |||
| - | def preprocess_input(image): | ||
| - | - image -= np.array((104, | ||
| - | + image = image / 255.0 | ||
| - | | ||
| - | ``` | ||
| After training model, we should convert pytorch model to onnx model. Create a python file written as follows and run. | After training model, we should convert pytorch model to onnx model. Create a python file written as follows and run. | ||
| Line 132: | Line 146: | ||
| ==== Get source code ==== | ==== Get source code ==== | ||
| - | Clone the source code form our [[gh> | + | Clone the source code from our [[gh> |
| ```shell | ```shell | ||
| Line 168: | Line 182: | ||
| $ bash build.sh | $ bash build.sh | ||
| - | # Run | + | # Run USB camera |
| + | $ cd install/ | ||
| + | $ ./ | ||
| + | |||
| + | # Run MIPI camera | ||
| $ cd install/ | $ cd install/ | ||
| - | $ ./ | + | $ ./ |
| ``` | ``` | ||
| - | '' | + | <WRAP info > |
| + | '' | ||
| + | </ | ||