可以同时使用 AVCaptureVideoDataOutput 和 AVCaptureMovieFileOutput 吗？

Posted 2023-02-23

【中文标题】可以同时使用 AVCaptureVideoDataOutput 和 AVCaptureMovieFileOutput 吗？

【英文标题】：Can use AVCaptureVideoDataOutput and AVCaptureMovieFileOutput at the same time?

【发布时间】：2011-06-24 01:33:51

【问题描述】：

我想用我的代码同时录制视频和抓取帧。

我使用AVCaptureVideoDataOutput 抓取帧，AVCaptureMovieFileOutput 用于视频录制。但不能同时工作，但单独工作时出现错误代码-12780。

我搜索了这个问题，但没有得到答案。有没有人有同样的经历或者解释一下？ 确实困扰了我一段时间。

谢谢。

【问题讨论】：

"可以使用 AVCaptureMovieFileOutput 将视频直接捕获到文件中。但是，该类没有可显示的数据，并且不能与 AVCaptureVideoDataOutput 同时使用。" i> 在这里找到：link .. 只是为了澄清问题的实际原因

【参考方案1】：

这是汤米回答的快速版本。

 // Set up the Capture Session 
 // Add the Inputs 
 // Add the Outputs


 var outputSettings = [
    AVVideoWidthKey : Int(640),
    AVVideoHeightKey : Int(480),
    AVVideoCodecKey : .h264
]

    var assetWriterInput = AVAssetWriterInput(mediaType: AVMediaTypeVideo,outputSettings: outputSettings)

    var pixelBufferAdaptor = AVAssetWriterInputPixelBufferAdaptor(assetWriterInput, sourcePixelBufferAttributes:
        [ kCVPixelBufferPixelFormatTypeKey : Int(kCVPixelFormatType_32BGRA)])


     var assetWriter = AVAssetWriter(url: URLFromSomwhere, fileType: AVFileTypeMPEG4 , error : Error )
         assetWriter.addInput(assetWriterInput)
         assetWriterInput.expectsMediaDataInRealTime = true
         assetWriter.startWriting()
         assetWriter.startSession(atSourceTime: kCMTimeZero)

    captureSession.startRunning()


  func captureOutput(_ captureOutput: AVCaptureOutput, didOutputSampleBuffer sampleBuffer: CMSampleBuffer, from connection: AVCaptureConnection) 

    var imageBuffer = CMSampleBufferGetImageBuffer(sampleBuffer)
    // a very dense way to keep track of the time at which this frame
    // occurs relative to the output stream, but it's just an example!
    var frameNumber: Int64 = 0

           if assetWriterInput.readyForMoreMediaData 
    pixelBufferAdaptor.appendPixelBuffer(imageBuffer, withPresentationTime: CMTimeMake(frameNumber, 25))
               
                frameNumber += 1   

      captureSession.stopRunning()
      assetWriter.finishWriting()

但我不能保证 100% 准确，因为我是 swift 新手。

【讨论】：

太棒了。这行得通。谢谢。但它需要更新到当前的语法。

【参考方案2】：

我无法回答提出的具体问题，但我已经成功地同时使用以下方法录制视频和抓取帧：

AVCaptureSession 和 AVCaptureVideoDataOutput 将帧路由到我自己的代码中 AVAssetWriter、AVAssetWriterInput 和 AVAssetWriterInputPixelBufferAdaptor 将帧写入 H.264 编码电影文件

这还没有调查音频。我最终从捕获会话中获得CMSampleBuffers，然后将它们推入像素缓冲区适配器。

编辑：所以我的代码看起来或多或少像，你在略过和忽略范围问题时没有问题：

/* to ensure I'm given incoming CMSampleBuffers */
AVCaptureSession *captureSession = alloc and init, set your preferred preset/etc;
AVCaptureDevice *captureDevice = default for video, probably;

AVCaptureDeviceInput *deviceInput = input with device as above, 
                                    and attach it to the session;

AVCaptureVideoDataOutput *output = output for 32BGRA pixel format, with me as the
                                   delegate and a suitable dispatch queue affixed.

/* to prepare for output; I'll output 640x480 in H.264, via an asset writer */
NSDictionary *outputSettings =
    [NSDictionary dictionaryWithObjectsAndKeys:

            [NSNumber numberWithInt:640], AVVideoWidthKey,
            [NSNumber numberWithInt:480], AVVideoHeightKey,
            AVVideoCodecH264, AVVideoCodecKey,

            nil];

AVAssetWriterInput *assetWriterInput = [AVAssetWriterInput 
                                   assetWriterInputWithMediaType:AVMediaTypeVideo
                                                  outputSettings:outputSettings];

/* I'm going to push pixel buffers to it, so will need a 
   AVAssetWriterPixelBufferAdaptor, to expect the same 32BGRA input as I've
   asked the AVCaptureVideDataOutput to supply */
AVAssetWriterInputPixelBufferAdaptor *pixelBufferAdaptor =
           [[AVAssetWriterInputPixelBufferAdaptor alloc] 
                initWithAssetWriterInput:assetWriterInput 
                sourcePixelBufferAttributes:
                     [NSDictionary dictionaryWithObjectsAndKeys:
                          [NSNumber numberWithInt:kCVPixelFormatType_32BGRA], 
                           kCVPixelBufferPixelFormatTypeKey,
                     nil]];

/* that's going to go somewhere, I imagine you've got the URL for that sorted,
   so create a suitable asset writer; we'll put our H.264 within the normal
   MPEG4 container */
AVAssetWriter *assetWriter = [[AVAssetWriter alloc]
                                initWithURL:URLFromSomwhere
                                fileType:AVFileTypeMPEG4
                                error:you need to check error conditions,
                                      this example is too lazy];
[assetWriter addInput:assetWriterInput];

/* we need to warn the input to expect real time data incoming, so that it tries
   to avoid being unavailable at inopportune moments */
assetWriterInput.expectsMediaDataInRealTime = YES;

... eventually ...

[assetWriter startWriting];
[assetWriter startSessionAtSourceTime:kCMTimeZero];
[captureSession startRunning];

... elsewhere ...

- (void)        captureOutput:(AVCaptureOutput *)captureOutput 
    didOutputSampleBuffer:(CMSampleBufferRef)sampleBuffer 
           fromConnection:(AVCaptureConnection *)connection

    CVImageBufferRef imageBuffer = CMSampleBufferGetImageBuffer(sampleBuffer);

    // a very dense way to keep track of the time at which this frame
    // occurs relative to the output stream, but it's just an example!
    static int64_t frameNumber = 0;
    if(assetWriterInput.readyForMoreMediaData)
        [pixelBufferAdaptor appendPixelBuffer:imageBuffer
                         withPresentationTime:CMTimeMake(frameNumber, 25)];
    frameNumber++;


... and, to stop, ensuring the output file is finished properly ...

[captureSession stopRunning];
[assetWriter finishWriting];

【讨论】：

您介意发布一个示例代码来说明如何做到这一点吗？你的真实生活业力会增加10倍！！！ :D

哦，有因果报应吗？然后添加了一些非常基本的示例代码！

感谢您的代码，我让它可以处理图像。给视频加声音怎么样，有什么线索吗？

设置输出 av 录制媒体的方向：developer.apple.com/library/ios/#qa/qa1744/_index.html#//…

你能整合声音吗？？ @ImranRaheem