我有兴趣在iPhone上添加文字（以字幕样式）到现有视频。一些解决这个在stackoverflow建议添加一个透明的UIView覆盖视频。这很好，但是，我不能将它保存为“新的和修改过的视频”。iOS在iPhone上逐帧修改视频（添加CC）
我看到的唯一解决方案就是取文本，从视频中取一帧，将文本添加到帧中，然后将带有文本的已修改帧推回到视频中，替换视频中的原始帧。
有谁知道如何从视频中取出一帧（我想我可以算出添加文本），然后如何将帧推回到视频中？如果你有一些想法，或知道教程，我将不胜感激。
你不需要在逐帧的基础上做到这一点。自iOS 4.0以来，AVFoundation支持字幕。
例如，您可以创建一个AVMutableComposition，然后在视频顶部添加一个字幕轨道。 AVMediaTypeSubtitle是字幕的类型（或用于隐藏字幕的AVMediaTypeClosedCaption）。然后，您可以将作品提供给玩家或AVAssetWriter。节省你所有的麻烦。
真棒 - 谢谢你，我会研究一下.. – geekyaleks 2013-03-18 16:39:07
我对你的解决方案感兴趣，但我没有找到如何使用“AVMediaTypeSubtitle”设置“字幕”的文本。你能举个例子吗？谢谢！ – lansher1985 2014-05-09 09:55:56
对于那些想要逐帧编辑电影的人，请使用AVReaderWriter。尽管它是一个OS X Apple示例代码，但AVFoundation在两个平台上都可用，几乎没有任何变化。
根据你所说的，这里有一个你可以学习的例子。
正如你可以从上面的图片中看到，你可以添加你想要的边界，叠加，字幕，什么视频。
http://www.raywenderlich.com/30200/avfoundation-tutorial-adding-overlays-and-animations-to-videos
您可以通过编辑框架AvFoundation视频......
下面是一个例子..
VAR的AssetTrack = AVAsset.assetWithURL（filePath1）为！ AVURLAsset var mutableComposition = AVMutableComposition（）
// Step - 1 pass url to avasset
// video
var compositionVideoTrack = AVMutableCompositionTrack()
compositionVideoTrack = mutableComposition.addMutableTrackWithMediaType(AVMediaTypeVideo, preferredTrackID: CMPersistentTrackID())
var assetVideoTrack = assetTrack.tracksWithMediaType(AVMediaTypeVideo)[0] as! AVAssetTrack
compositionVideoTrack.insertTimeRange(CMTimeRangeMake(kCMTimeZero , assetTrack.duration), ofTrack: assetVideoTrack, atTime: kCMTimeZero, error: nil)
// audio
var compositionAudioTrack = AVMutableCompositionTrack()
compositionAudioTrack = mutableComposition.addMutableTrackWithMediaType(AVMediaTypeAudio, preferredTrackID: CMPersistentTrackID())
var assetAudioTrack = assetTrack.tracksWithMediaType(AVMediaTypeAudio)[0] as! AVAssetTrack
compositionAudioTrack.insertTimeRange(CMTimeRangeMake(kCMTimeZero , assetTrack.duration), ofTrack: assetAudioTrack, atTime: kCMTimeZero, error: nil)
var videoAssetOrientation_: UIImageOrientation = .Up
var isVideoAssetPortrait_: Bool = false
var videoTransform: CGAffineTransform = assetTrack.preferredTransform
var videosize = CGSize()
videosize = assetVideoTrack.naturalSize
var parentLayer = CALayer()
var videoLayer = CALayer()
var textLayer = CALayer()
parentLayer.frame = CGRectMake(0, 0, videosize.width, videosize.height)
videoLayer.frame = CGRectMake(0, 0, videosize.width, videosize.height)
textLayer.frame = CGRectMake(0, 0, videosize.width, videosize.height)
parentLayer.addSublayer(videoLayer)
if drawingView.image != nil
{
var drawingLayer = CALayer()
drawingLayer.frame = CGRectMake(0, 0, videosize.width, videosize.height)
drawingLayer.contents = drawingView.image.CGImage
var image = UIImage()
parentLayer.addSublayer(drawingLayer)
}
var textV = UIView()
textV.backgroundColor = UIColor.clearColor()
textV.layer.backgroundColor = UIColor.clearColor().CGColor
textV.frame = CGRectMake(self.captureView.frame.size.width, 0, self.captureView.frame.size.width, self.captureView.frame.size.height)
var textL = UILabel()
textL = textShowOnPreview
textV.addSubview(textL)
if textL != ""
{
UIGraphicsBeginImageContext(textV.bounds.size)
textV.layer.renderInContext(UIGraphicsGetCurrentContext())
var image1: UIImage = UIGraphicsGetImageFromCurrentImageContext()
UIGraphicsEndImageContext()
if (TextAnimation == "")
{
textLayer.contents = image1.CGImage
parentLayer.addSublayer(textLayer)
}
else if (TextAnimation == "flip")
{
var overlayer1 = CALayer()
overlayer1.backgroundColor = UIColor.clearColor().CGColor
let screenSize: CGRect = UIScreen.mainScreen().bounds
overlayer1.contents = image1.CGImage
overlayer1.masksToBounds = true
overlayer1.frame = CGRectMake(videosize.width/2-300, videosize.height/2 - 400, videosize.width,videosize.width);
var animation : CABasicAnimation = CABasicAnimation(keyPath: "transform.rotation")
animation.duration=5.0;
animation.repeatCount=5;
animation.autoreverses = true;
// rotate from 0 to 360
animation.fromValue = 0
animation.toValue = (2.0 * M_PI);
animation.beginTime = AVCoreAnimationBeginTimeAtZero;
overlayer1.addAnimation(animation, forKey:"rotation")
[parentLayer .addSublayer(overlayer1)]
}
else if (TextAnimation == "fade")
{
//opacity
var overlayer1 = CALayer()
overlayer1.backgroundColor = UIColor.clearColor().CGColor
overlayer1.contents = image1.CGImage
overlayer1.masksToBounds = true
overlayer1.frame = CGRectMake(videosize.width/2 - 300, videosize.height/2 - 100 , videosize.width+20, videosize.width);
var animation : CABasicAnimation = CABasicAnimation(keyPath: "transform.scale")
animation.duration = 2.0;
animation.repeatCount = 3;
animation.autoreverses = true;
// rotate from 0 to 360
animation.fromValue = 0.5;
animation.toValue = 1.0;
animation.beginTime = AVCoreAnimationBeginTimeAtZero;
overlayer1.addAnimation(animation, forKey:"scale")
[parentLayer .addSublayer(overlayer1)]
}
else if (TextAnimation == "bounce")
{
var overlayer1 = CALayer()
var bounce : CABasicAnimation = CABasicAnimation (keyPath:"position.y");
overlayer1.backgroundColor = UIColor.clearColor().CGColor
overlayer1.contents = image1.CGImage
overlayer1.masksToBounds = true
overlayer1.frame = CGRectMake(videosize.width/2 - 300, videosize.height/2 - 100 , videosize.width, videosize.width);
bounce.duration = 1.0;
bounce.fromValue = overlayer1.frame.origin.y
bounce.toValue = overlayer1.frame.origin.y - 100
bounce.repeatCount = 10
bounce.autoreverses = true;
overlayer1.addAnimation(bounce, forKey: "y")
var animation = CABasicAnimation(keyPath: "transform.scale")
animation.toValue = NSNumber(float: 0.9)
animation.duration = 1.0
animation.repeatCount = 10;
animation.autoreverses = true
overlayer1.addAnimation(animation, forKey: nil)
[parentLayer .addSublayer(overlayer1)]
}
}
var mutableVideoComposition = AVMutableVideoComposition()
mutableVideoComposition.frameDuration = CMTimeMake(1, 30)
mutableVideoComposition.renderSize = videosize
mutableVideoComposition.animationTool = AVVideoCompositionCoreAnimationTool(postProcessingAsVideoLayer: videoLayer, inLayer: parentLayer)
var passThroughInstruction = AVMutableVideoCompositionInstruction()
passThroughInstruction.timeRange = CMTimeRangeMake(kCMTimeZero, mutableComposition.duration)
var passThroughLayerInstruction = AVMutableVideoCompositionLayerInstruction()
// video
var assestVideoMutableCompositionVideo = mutableComposition.tracksWithMediaType(AVMediaTypeVideo)[0] as! AVAssetTrack
passThroughLayerInstruction = AVMutableVideoCompositionLayerInstruction(assetTrack: assestVideoMutableCompositionVideo)
if isVideoAssetPortrait_ == false
{
var FirstAssetScaleFactor: CGAffineTransform = CGAffineTransformMakeScale(1, 1)
passThroughLayerInstruction.setTransform(CGAffineTransformConcat(assetVideoTrack.preferredTransform, FirstAssetScaleFactor), atTime: kCMTimeZero)
}
passThroughInstruction.layerInstructions = [passThroughLayerInstruction]
mutableVideoComposition.instructions = [passThroughInstruction]
let documentsURL = NSFileManager.defaultManager().URLsForDirectory(.DocumentDirectory, inDomains: .UserDomainMask)[0] as! NSURL
let filePath = documentsURL.URLByAppendingPathComponent("NewWatermarkedVideo.mov") as NSURL
var fileManager:NSFileManager = NSFileManager.defaultManager()
fileManager.removeItemAtURL(filePath, error: nil)
var exporter: AVAssetExportSession = AVAssetExportSession(asset: mutableComposition, presetName: AVAssetExportPresetMediumQuality)
exporter.videoComposition = mutableVideoComposition
exporter.outputFileType = AVFileTypeQuickTimeMovie
exporter.outputURL = filePath
exporter.shouldOptimizeForNetworkUse = true
self.captureView.addSubview(textShowOnPreview)
exporter.exportAsynchronouslyWithCompletionHandler({() -> Void in
println(exporter.status)
if exporter.status == AVAssetExportSessionStatus.Completed
{
dispatch_async(dispatch_get_main_queue(), {() -> Void in
MBProgressHUD.hideAllHUDsForView(self.view, animated: true)
self.topicSelectedImage.highlighted = false
self.timelineSelectedImage.highlighted = false
self.selectCat = ""
self.postView.hidden = false
})
println("Completed")
self.mediaData = NSData(contentsOfURL:filePath, options: nil, error: nil)!
var err: NSError? = nil
var asset = AVURLAsset(URL: filePath, options: nil)
var imgGenerator = AVAssetImageGenerator(asset: asset)
var cgImage = imgGenerator.copyCGImageAtTime(CMTimeMake(0, 30), actualTime: nil, error: &err)
var uiImage = UIImage(CGImage: cgImage)!
self.videoThumbData = UIImageJPEGRepresentation(uiImage, 0.1)
var assetTrack = AVAsset.assetWithURL(filePath) as! AVURLAsset
self.videoTime = Int(CMTimeGetSeconds(assetTrack.duration)) + 3
println(self.videoTime)
}
else if exporter.status == AVAssetExportSessionStatus.Cancelled
{
}
else if exporter.status == AVAssetExportSessionStatus.Failed
{
}
})
你是否浏览了AVFoundation框架文档？ – zoul 2013-03-18 15:06:31
只要每次播放视频时，为什么要将其保存为修改过的视频呢？编辑视频是CPU密集型的，因此移动设备上的电池耗尽。 – codeghost 2013-03-18 15:10:21
iPhone已经具有视频编辑功能，以及iPhone的完整iMovie套件，所以我不知道如何添加文本框架是更多的税收.... – geekyaleks 2013-03-18 15:29:45