AVCaptureMetadataOutput如何判断人脸检测何时丢失人脸且帧为空

import AVFoundation

class FaceDetectionViewController: UIViewController, AVCaptureMetadataOutputObjectsDelegate {
    var captureSession: AVCaptureSession!
    var metadataOutput: AVCaptureMetadataOutput!
    
    override func viewDidLoad() {
        super.viewDidLoad()
        
        // 创建会话
        captureSession = AVCaptureSession()
        
        // 获取摄像头设备
        guard let videoDevice = AVCaptureDevice.default(for: .video) else {
            print("无法获取摄像头设备")
            return
        }
        
        // 创建输入对象
        guard let videoInput = try? AVCaptureDeviceInput(device: videoDevice) else {
            print("无法创建输入对象")
            return
        }
        
        // 将输入对象添加到会话中
        if captureSession.canAddInput(videoInput) {
            captureSession.addInput(videoInput)
        }
        
        // 创建输出对象
        metadataOutput = AVCaptureMetadataOutput()
        
        // 将输出对象添加到会话中
        if captureSession.canAddOutput(metadataOutput) {
            captureSession.addOutput(metadataOutput)
        }
        
        // 设置输出对象的代理
        metadataOutput.setMetadataObjectsDelegate(self, queue: DispatchQueue.main)
        
        // 设置输出对象的元数据类型为人脸检测
        metadataOutput.metadataObjectTypes = [.face]
        
        // 开始会话
        captureSession.startRunning()
    }
    
    // AVCaptureMetadataOutputObjectsDelegate方法
    func metadataOutput(_ output: AVCaptureMetadataOutput, didOutput metadataObjects: [AVMetadataObject], from connection: AVCaptureConnection) {
        if metadataObjects.isEmpty {
            // 丢失了人脸
            print("丢失了人脸")
        } else {
            // 检测到了人脸
            print("检测到了人脸")
        }
    }
}