问题描述
我目前正在开发一款iOS应用程序,将CoreImage应用于相机Feed以拍摄照片和视频,而且我遇到了一些麻烦。
I'm currently developing an iOS app that applies CoreImage to the camera feed in order to take photos and videos, and I've run into a bit of snag.
到目前为止,我一直在使用 AVCaptureVideoDataOutput 获取样本缓冲区并使用CoreImage对其进行操作,然后显示简单的预览,并使用它来捕获照片并保存它们。
Up till now I've been using AVCaptureVideoDataOutput to obtain the sample buffers and manipulate them with CoreImage, and then displayed a simple preview, as well as using it to capture photos and saving them.
当我尝试实施视频录制时,将我从 AVCaptureVideoDataOutput 收到的视频写入SampleBuffers,它的速度非常慢帧速率(可能是因为正在进行的其他图像相关处理)。
When I tried to implement Video Recording, by writing the SampleBuffers to a video as I received them from the AVCaptureVideoDataOutput, it had a very slow frame rate (probably because of the other image relating processing that was going on).
所以我想知道,是否有可能拥有 AVCaptureVideoDataOutput 和AVCaptureMoveFileOutput同时在同一个AVCaptureSession上进行?
So I was wondering, is it possible to have an AVCaptureVideoDataOutput and a AVCaptureMoveFileOutput going on the same AVCaptureSession simultaneously?
我快速给了它,发现当我添加额外输出时,我的AVCaptureVideoDataOutput停止接收信息。
I gave it a quick go, and found that when I added the extra output, my AVCaptureVideoDataOutput stopped receiving information.
如果我可以使它工作,我希望这意味着我可以简单地使用第二个输出以高帧率记录视频,并进行后期处理用户停止录制后的视频。
If I can get it working, I'm hoping it means that I can simply use the 2nd output to record video at high frame rates, and do post-processing on the video after the user has stopped recording.
我们将非常感谢任何帮助。
Any help will be greatly appreciated.
推荐答案
这比你想象的容易。
参见:
- 使用AVCaptureVideoDataOutput捕获数据。
- 在录制之前创建一个新的调度队列,例如。 recordingQueue:
recordingQueue =
dispatch_queue_create(电影录制队列,
DISPATCH_QUEUE_SERIAL); -
在captureOutput:didOutputSampleBuffer:fromConnection:委托
方法,捕获samplebuffer,保留它,并在记录
队列中,将其写入文件:
- Capture data using AVCaptureVideoDataOutput.
- Create a new dispatch queue before recording, eg. recordingQueue:
recordingQueue =dispatch_queue_create("Movie Recording Queue",DISPATCH_QUEUE_SERIAL);
In the captureOutput:didOutputSampleBuffer:fromConnection: delegate method, capture the samplebuffer, retain it, and in the recording queue, write it to the file:
-(void)captureOutput:(AVCaptureOutput *)captureOutput didOutputSampleBuffer:(CMSampleBufferRef)sampleBuffer fromConnection:(AVCaptureConnection *)connection {
CFRetain(sampleBuffer);
dispatch_async(recordingQueue, ^{
if (assetWriter) {
if (connection == videoConnection) {
[self writeSampleBuffer:sampleBuffer ofType:AVMediaTypeVideo];
} else if (connection == audioConnection) {
[self writeSampleBuffer:sampleBuffer ofType:AVMediaTypeAudio];
}
}
CFRelease(sampleBuffer);
});
}
- (void) writeSampleBuffer:(CMSampleBufferRef)sampleBuffer ofType:(NSString *)mediaType
{
CMTime presentationTime = CMSampleBufferGetPresentationTimeStamp(sampleBuffer);
if ( assetWriter.status == AVAssetWriterStatusUnknown ) {
if ([assetWriter startWriting]) {
[assetWriter startSessionAtSourceTime:presentationTime];
} else {
NSLog(@"Error writing initial buffer");
}
}
if ( assetWriter.status == AVAssetWriterStatusWriting ) {
if (mediaType == AVMediaTypeVideo) {
if (assetWriterVideoIn.readyForMoreMediaData) {
if (![assetWriterVideoIn appendSampleBuffer:sampleBuffer]) {
NSLog(@"Error writing video buffer");
}
}
}
else if (mediaType == AVMediaTypeAudio) {
if (assetWriterAudioIn.readyForMoreMediaData) {
if (![assetWriterAudioIn appendSampleBuffer:sampleBuffer]) {
NSLog(@"Error writing audio buffer");
}
}
}
}
}
这篇关于具有多个输出的AVCaptureSession?的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持!