CaptureManager.swift

class CaptureManager : NSObject, AVCaptureVideoDataOutputSampleBufferDelegate {
    override init() {
        super.init()
        startCaptureSession()
    }
    
    deinit {
        stopCaptureSession()
    }
    
    func setCameraCallback(callback: @escaping (CGImage) -> Void) {
        cameraCallback = callback
    }
    
    func captureOutput(_ output: AVCaptureOutput, didOutput sampleBuffer: CMSampleBuffer, from connection: AVCaptureConnection) {
        if output == videoOutput {
            // Keep track of frames
            cameraInputFrameCount += 1
            
            NSLog("Camera capture frame \(cameraInputFrameCount)")
            
            if let image = cgImageFromSampleBuffer(sampleBuffer) {
                DispatchQueue.main.async {
                    self.cameraCallback(image)
                }
            }
        }
    }
    
    private let mediaQueue = DispatchQueue(label: "media-queue")
    
    // Camera and microphone capture
    private var captureSession: AVCaptureSession?
    private var cameraInputFrameCount = 0
    
    private var videoOutput: AVCaptureOutput?
    private var audioOutput: AVCaptureOutput?
    
    private var cameraCallback: (CGImage) -> Void = { (image) in }
    
    private func startCaptureSession() {
        let captureSession = AVCaptureSession()
        captureSession.beginConfiguration()
        
        // Video
        if
            let videoDevice = AVCaptureDevice.default(for: .video),
            let videoInput = try? AVCaptureDeviceInput(device: videoDevice)
        {
            if captureSession.canAddInput(videoInput) {
                captureSession.addInput(videoInput)
                
                let videoOutput = AVCaptureVideoDataOutput()
                videoOutput.setSampleBufferDelegate(self, queue: mediaQueue)
                videoOutput.videoSettings = [kCVPixelBufferPixelFormatTypeKey as String: Int(kCVPixelFormatType_32BGRA)]

                if captureSession.canAddOutput(videoOutput) {
                    captureSession.addOutput(videoOutput)
                    self.videoOutput = videoOutput
                }
            }
        }
        
        captureSession.commitConfiguration()
        captureSession.startRunning()
        
        self.captureSession = captureSession
    }
    
    private func stopCaptureSession() {
        captureSession?.stopRunning()
        captureSession = nil
    }
    
    private func cgImageFromSampleBuffer(_ sampleBuffer: CMSampleBuffer) -> CGImage? {
        // Get the CVImageBuffer (pixel buffer) from the sample buffer
        guard let imageBuffer = CMSampleBufferGetImageBuffer(sampleBuffer) else {
            print("Failed to get image buffer from sample buffer")
            return nil
        }
        
        CVPixelBufferLockBaseAddress(imageBuffer, .readOnly)
        defer {
            CVPixelBufferUnlockBaseAddress(imageBuffer, .readOnly)
        }
        
        // Create a CIImage from the pixel buffer
        let ciImage = CIImage(cvPixelBuffer: imageBuffer)
        
        // Create a CIContext to render the CIImage to a CGImage
        let context = CIContext()
        
        // Convert CIImage to CGImage
        let cgImage = context.createCGImage(ciImage, from: ciImage.extent)
        
        return cgImage
    }

}