From 4dd4b7c43e1934c27172083b09d5c74928ebf093 Mon Sep 17 00:00:00 2001 From: dengwenjie <775747758@qq.com> Date: Mon, 14 Apr 2025 14:02:59 +0800 Subject: [PATCH] =?UTF-8?q?=E6=96=B0=E5=A2=9E=E7=9B=AE=E6=A0=87=E6=A3=80?= =?UTF-8?q?=E6=B5=8B=E5=8A=9F=E8=83=BD?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- README.md | 26 ++++++++++++++++++++++++++ 1 file changed, 26 insertions(+) diff --git a/README.md b/README.md index 6b35284..8a3a0ef 100644 --- a/README.md +++ b/README.md @@ -95,7 +95,9 @@ 1.0.8 ``` + ### 3、人脸检测运行流程 +

@@ -161,6 +163,30 @@ https://gitee.com/dengwenjie/SmartJavaAI-Demo └── 📄[ObjectDetection.java](https://gitee.com/dengwenjie/SmartJavaAI-Demo/blob/master/src/main/java/smartai/examples/objectdetection/ObjectDetection.java) *(目标检测示例代码)* +### 5、运行设备 +默认将使用CPU运行模型,如果需要使用GPU,需要在创建模型接口中手动指定运行设备 + +GPU中运行需要满足如下条件: + +(1)安装CUDA v12.4及cuDNN v8.9.7 + +(2)将缓存目录加入到环境变量中,并删除其他cuda环境变量 + +​ windows配置如下: + +

+ +

+ + +(3)示例代码: + +```java +FaceModelConfig config = new FaceModelConfig(); +config.setModelEnum(FaceModelEnum.RETINA_FACE);//人脸模型 +config.setDevice(DeviceEnum.GPU);//指定GPU +FaceModel faceModel = FaceModelFactory.getInstance().getModel(config); +``` ## 🙏 鸣谢