我试图创建一个简单的视频应用程序(从ios 4设备加载现有视频文件,使用直接像素访问进行编辑并以不同的名称进行保存)。iOS 4:基于AVAsset的应用程序的视频质量
我设法加载,编辑和保存我的电影文件在真实设备(ipod 4g)。我唯一的问题是与电影质量(原始与编辑之一)有关。我不知道我在做什么错,但输出文件的质量与输入文件相比非常糟糕。
下面你可以找到我怎么装我的电影:
// // *** tmp file ***
NSURL *movieUrl = [info objectForKey:@"UIImagePickerControllerMediaURL"];
NSLog(@"picker controller movie url: %@", [movieUrl absoluteString]);
## picker controller movie url: /private/var/mobile/Applications/6253D1-8C0-41-B780-638250817/tmp//trim.2q03uz.MOV
AVURLAsset *movieAsset = [[AVURLAsset alloc] initWithURL:movieUrl options:nil];
CGSize size = [movieAsset naturalSize];
NSLog(@"movie asset natual size: size.width = %f size.height = %f", size.width, size.height);
## movie asset natual size: size.width = 224.000000 size.height = 128.000000
// *** ASSET READER ***
AVAssetReader *assetReader = [[AVAssetReader alloc] initWithAsset:movieAsset error:&error];
NSArray* videoTracks = [movieAsset tracksWithMediaType:AVMediaTypeVideo];
// asset track
AVAssetTrack* videoTrack = [videoTracks objectAtIndex:0];
NSDictionary* dictionary = [NSDictionary dictionaryWithObject:[NSNumber numberWithInt:kCVPixelFormatType_420YpCbCr8BiPlanarVideoRange]
forKey:(id)kCVPixelBufferPixelFormatTypeKey];
NSLog(@"video track %f %f %d", [videoTrack naturalSize].width, [videoTrack naturalSize].height, [videoTrack nominalFrameRate]);
## video track 224.000000 128.000000 0
/*
also tried
kCVPixelFormatType_32BGRA
kCVPixelFormatType_420YpCbCr8BiPlanarVideoRange
*/
// asset reader track output
AVAssetReaderTrackOutput* assetReaderOutput = [[AVAssetReaderTrackOutput alloc] initWithTrack:videoTrack
outputSettings:dictionary];
if(![assetReader canAddOutput:assetReaderOutput])
NSLog(@"unable to add reader output");
else
[assetReader addOutput:assetReaderOutput];
// *** ASSET WRITER ***
AVAssetWriter *videoWriter = [[AVAssetWriter alloc] initWithURL:exportURL
fileType:AVFileTypeQuickTimeMovie error:&error];
NSParameterAssert(videoWriter);
NSLog(@"asset writer %d %d", [videoWriter status], [error code]);
NSDictionary *videoCompressionProps = [NSDictionary dictionaryWithObjectsAndKeys:
[NSNumber numberWithDouble:128.0*1024.0], AVVideoAverageBitRateKey,
nil];
NSDictionary *videoSettings = [NSDictionary dictionaryWithObjectsAndKeys:
AVVideoCodecH264, AVVideoCodecKey,
[NSNumber numberWithInt:size.width], AVVideoWidthKey,
[NSNumber numberWithInt:size.height], AVVideoHeightKey,
//videoCompressionProps, AVVideoCompressionPropertiesKey,<- no difference at all
nil];
AVAssetWriterInput* writerInput = [[AVAssetWriterInput
assetWriterInputWithMediaType:AVMediaTypeVideo
outputSettings:videoSettings] retain];
// set preffered transform for output video
writerInput.transform = [videoTrack preferredTransform];
NSParameterAssert(writerInput);
NSParameterAssert([videoWriter canAddInput:writerInput]);
[videoWriter addInput:writerInput];
writerInput.expectsMediaDataInRealTime = NO;
然后,我只是读取下一个样本缓冲区,修改像素数据,并用不同的名称保存它。这已经工作,但输出质量是如此糟糕......
请大家给我一些建议:-)。
非常好的努力。\ –