[Model] Add FSANet model (#448)

* add yolov5cls * fixed bugs * fixed bugs * fixed preprocess bug * add yolov5cls readme * deal with comments * Add YOLOv5Cls Note * add yolov5cls test * add rvm support * support rvm model * add rvm demo * fixed bugs * add rvm readme * add TRT support * add trt support * add rvm test * add EXPORT.md * rename export.md * rm poros doxyen * deal with comments * deal with comments * add rvm video_mode note * add fsanet * fixed bug * update readme * fixed for ci * deal with comments * deal with comments * deal with comments Co-authored-by: Jason <jiangjiajun@baidu.com> Co-authored-by: DefTruth <31974251+DefTruth@users.noreply.github.com>
2025-10-11 03:20:20 +08:00 · 2022-11-04 11:00:35 +08:00
parent ce828ecb38
commit 7150e6405c
31 changed files with 922 additions and 22 deletions
--- a/examples/vision/headpose/fsanet/python/README.md
+++ b/examples/vision/headpose/fsanet/python/README.md
@@ -0,0 +1,67 @@
+# FSANet Python部署示例
+
+在部署前，需确认以下两个步骤
+
+- 1. 软硬件环境满足要求，参考[FastDeploy环境要求](../../../../../docs/cn/build_and_install/download_prebuilt_libraries.md)  
+- 2. FastDeploy Python whl包安装，参考[FastDeploy Python安装](../../../../../docs/cn/build_and_install/download_prebuilt_libraries.md)
+
+本目录下提供`infer.py`快速完成FSANet在CPU/GPU，以及GPU上通过TensorRT加速部署的示例，保证 FastDeploy 版本 >= 0.6.0 支持FSANet模型。执行如下脚本即可完成
+
+```bash
+#下载部署示例代码
+git clone https://github.com/PaddlePaddle/FastDeploy.git
+cd FastDeploy/examples/vision/headpose/fsanet/python
+
+# 下载FSANet模型文件和测试图片
+## 原版ONNX模型
+wget https://bj.bcebos.com/paddlehub/fastdeploy/fsanet-var.onnx
+wget https://bj.bcebos.com/paddlehub/fastdeploy/headpose_input.png
+# CPU推理
+python infer.py --model fsanet-var.onnx --image headpose_input.png --device cpu
+# GPU推理
+python infer.py --model fsanet-var.onnx --image headpose_input.png --device gpu
+# TRT推理
+python infer.py --model fsanet-var.onnx --image headpose_input.png --device gpu --backend trt
+```
+
+运行完成可视化结果如下图所示
+
+<div width="520">
+<img width="500" height="514" float="left" src="https://user-images.githubusercontent.com/19977378/198279932-3eee424e-98a2-4249-bdeb-0f79127cbc9d.png">
+</div>
+
+## FSANet Python接口
+
+```python
+fd.vision.headpose.FSANet(model_file, params_file=None, runtime_option=None, model_format=ModelFormat.ONNX)
+```
+
+FSANet 模型加载和初始化，其中model_file为导出的ONNX模型格式
+
+**参数**
+
+> * **model_file**(str): 模型文件路径
+> * **params_file**(str): 参数文件路径，当模型格式为ONNX格式时，此参数无需设定
+> * **runtime_option**(RuntimeOption): 后端推理配置，默认为None，即采用默认配置
+> * **model_format**(ModelFormat): 模型格式，默认为ONNX
+### predict函数
+
+> ```python
+> FSANet.predict(input_image)
+> ```
+>
+> 模型预测结口，输入图像直接输出头部姿态预测结果。
+>
+> **参数**
+>
+> > * **input_image**(np.ndarray): 输入数据，注意需为HWC，BGR格式
+> **返回**
+>
+> > 返回`fastdeploy.vision.HeadPoseResult`结构体，结构体说明参考文档[视觉模型预测结果](../../../../../docs/api/vision_results/)
+
+## 其它文档
+
+- [FSANet 模型介绍](..)
+- [FSANet C++部署](../cpp)
+- [模型预测结果说明](../../../../../docs/api/vision_results/)
+- [如何切换模型推理后端引擎](../../../../../docs/cn/faq/how_to_change_backend.md)
--- a/examples/vision/headpose/fsanet/python/infer.py
+++ b/examples/vision/headpose/fsanet/python/infer.py
@@ -0,0 +1,88 @@
+import fastdeploy as fd
+import cv2
+import os
+
+
+def parse_arguments():
+    import argparse
+    import ast
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--model", required=True, help="Path of FSANet model.")
+    parser.add_argument("--image", type=str, help="Path of test image file.")
+    parser.add_argument(
+        "--device",
+        type=str,
+        default='cpu',
+        help="Type of inference device, support 'cpu' or 'gpu'.")
+    parser.add_argument(
+        "--backend",
+        type=str,
+        default="default",
+        help="inference backend, default, ort, ov, trt, paddle, paddle_trt.")
+    parser.add_argument(
+        "--enable_trt_fp16",
+        type=ast.literal_eval,
+        default=False,
+        help="whether enable fp16 in trt/paddle_trt backend")
+    return parser.parse_args()
+
+
+def build_option(args):
+    option = fd.RuntimeOption()
+    device = args.device
+    backend = args.backend
+    enable_trt_fp16 = args.enable_trt_fp16
+    if device == "gpu":
+        option.use_gpu()
+        if backend == "ort":
+            option.use_ort_backend()
+        elif backend == "paddle":
+            option.use_paddle_backend()
+        elif backend in ["trt", "paddle_trt"]:
+            option.use_trt_backend()
+            option.set_trt_input_shape("input", [1, 3, 64, 64])
+            if backend == "paddle_trt":
+                option.enable_paddle_to_trt()
+            if enable_trt_fp16:
+                option.enable_trt_fp16()
+        elif backend == "default":
+            return option
+        else:
+            raise Exception(
+                "While inference with GPU, only support default/ort/paddle/trt/paddle_trt now, {} is not supported.".
+                format(backend))
+    elif device == "cpu":
+        if backend == "ort":
+            option.use_ort_backend()
+        elif backend == "ov":
+            option.use_openvino_backend()
+        elif backend == "paddle":
+            option.use_paddle_backend()
+        elif backend == "default":
+            return option
+        else:
+            raise Exception(
+                "While inference with CPU, only support default/ort/ov/paddle now, {} is not supported.".
+                format(backend))
+    else:
+        raise Exception(
+            "Only support device CPU/GPU now, {} is not supported.".format(
+                device))
+
+    return option
+
+
+args = parse_arguments()
+
+# 配置runtime，加载模型
+runtime_option = build_option(args)
+model = fd.vision.headpose.FSANet(args.model, runtime_option=runtime_option)
+
+# for image
+im = cv2.imread(args.image)
+result = model.predict(im.copy())
+print(result)
+# 可视化结果
+vis_im = fd.vision.vis_headpose(im, result)
+cv2.imwrite("visualized_result.jpg", vis_im)
+print("Visualized result save in ./visualized_result.jpg")