说明书

实验八/README.MD

@@ -0,0 +1,35 @@
+# 简单的人脸识别
+
+## 玩法
+
+###  配置环境
+
+#### 1.安装requirements.txt
+
+```bash
+pip install -r requirements.txt
+```
+
+#### 2.下载dlib模型
+
+只需执行我的自动下载脚本即可，需`魔法上网`
+
+```bash
+./init_enviorment.sh
+```
+
+```powershell
+.\init_enviorment.ps1
+```
+
+### 替换数据集
+
+- 替换`cache\dataset.mp4`
+
+- 录一段你想要识别为True的人脸的视频
+
+- 然后训练
+
+### 测试图片路径
+
+测试图片在`cache`文件夹内，一个`positive.jpg`，一个`negative.jpg`，分别是正负样本

实验八/init_enviorment.ps1

@@ -0,0 +1,10 @@
+mkdir models
+mkdir cache
+wget https://github.com/davisking/dlib-models/raw/refs/heads/master/shape_predictor_68_face_landmarks_GTX.dat.bz2
+wget https://github.com/davisking/dlib-models/raw/refs/heads/master/dlib_face_recognition_resnet_model_v1.dat.bz2
+Move-Item dlib_face_recognition_resnet_model_v1.dat.bz2 models/
+Move-Item shape_predictor_68_face_landmarks_GTX.dat.bz2 models/
+Set-Location models
+7z x dlib_face_recognition_resnet_model_v1.dat.bz2
+7z x shape_predictor_68_face_landmarks_GTX.dat.bz2
+Write-Output initialized enviorment!

实验八/init_enviorment.sh

@@ -0,0 +1,11 @@
+#!/bin/bash
+wget https://github.com/davisking/dlib-models/raw/refs/heads/master/shape_predictor_68_face_landmarks_GTX.dat.bz2
+wget https://github.com/davisking/dlib-models/raw/refs/heads/master/dlib_face_recognition_resnet_model_v1.dat.bz2
+mkdir cache
+mkdir models
+mv dlib_face_recognition_resnet_model_v1.dat.bz2 models/
+mv shape_predictor_68_face_landmarks_GTX.dat.bz2 models/
+cd models
+bzip2 -d dlib_face_recognition_resnet_model_v1.dat.bz2
+bzip2 -d shape_predictor_68_face_landmarks_GTX.dat.bz2
+echo initialized enviorment!