将视频与叠加文本合并

Question

在

Swift

中，我有一个

iOS

数组。我通过一个函数将其传递，将视频资源拼接/合并到一个最终视频中。对于每个子视频，我的目标是覆盖位于视频帧中心的文本。

到目前为止，我已经实现了这个

的半工作版本

基于我以前的帖子，但遇到了以下问题我无法理解......

每个视频的叠加文本在最终合并视频的整个持续时间内相互重叠。每个文本应与其子视频的持续时间、开始/结束时间相对应。
视频以纵向、垂直方向录制，为了保持一致，我认为将
AVURLAsset
```
 
```
AVMutableVideoCompositionLayerInstruction
```
 设置为相应的 
```
.preferredTransform
```
 视频 
```
AVAssetTrack
```
 可以做到这一点，但事实并非如此。视频方向为 -90 度，屏幕下半部分为黑色。
```

我最初是通过

.preferredTransform

 播放最终视频资源的，它是

AVAssetExportSession

 属性。但我注意到，只有当我从

.asset

的

AVAsset

 属性创建新的

AVAssetExportSession

 时，才会显示文本叠加。这需要

.outputUrl

 在播放视频之前完全完成，而之前当我从

AVAssetExportSession

 属性播放时，最终视频会立即播放已加载的内容，而无需等待整个内容加载（但同样，没有文本）覆盖）。有什么方法可以在渲染文本叠加时返回到原始状态吗？

我尝试遵循

现有答案和Ray Wenderlich的教程，但没有成功。

任何指导都会

非常赞赏..

.asset

Answer 1

我希望有一些时间来回顾这篇文章，但这是下面的代码。

这是创建视频的代码。

func merge(videos: [AVURLAsset], completion: @escaping (_ url: URL, _ asset: AVAssetExportSession)->()) {

let videoComposition = AVMutableComposition()
var lastTime: CMTime = .zero

var maxVideoSize = CGSize.zero

guard let videoCompositionTrack = videoComposition.addMutableTrack(withMediaType: .video, preferredTrackID: Int32(kCMPersistentTrackID_Invalid)),
      let audioCompositionTrack = videoComposition.addMutableTrack(withMediaType: .audio, preferredTrackID: Int32(kCMPersistentTrackID_Invalid)) else { return }

let mainComposition = AVMutableVideoComposition()

let mainParentLayer = CALayer()
let mainVideoLayer = CALayer()
mainParentLayer.frame = CGRect(x: 0, y: 0, width: maxVideoSize.width, height: maxVideoSize.height)
mainVideoLayer.frame = CGRect(x: 0, y: 0, width: maxVideoSize.width, height: maxVideoSize.height)

mainParentLayer.addSublayer(mainVideoLayer)

var instructions = [AVMutableVideoCompositionInstruction]()

for video in videos {
    
    if let videoTrack = video.tracks(withMediaType: .video)[safe: 0], let text = savedTexts[video.url] {
                        
        videoCompositionTrack.preferredTransform = videoTrack.preferredTransform
        
        do {
            try videoCompositionTrack.insertTimeRange(CMTimeRangeMake(start: .zero, duration: video.duration), of: videoTrack, at: lastTime)
            
            if let audioTrack = video.tracks(withMediaType: .audio)[safe: 0] {
                try audioCompositionTrack.insertTimeRange(CMTimeRangeMake(start: .zero, duration: video.duration), of: audioTrack, at: lastTime)
            }
            
            lastTime = CMTimeAdd(lastTime, video.duration)
            
            let videoSize = videoTrack.naturalSize.applying(videoTrack.preferredTransform)
            let videoRect = CGRect(x: 0, y: 0, width: abs(videoSize.width), height: abs(videoSize.height))
            maxVideoSize = CGSize(width: max(maxVideoSize.width, videoRect.width), height: max(maxVideoSize.height, videoRect.height))
            
            let textLayer = CATextLayer()
            textLayer.string = text
            textLayer.foregroundColor = UIColor.white.cgColor
            textLayer.font = UIFont(name: "Helvetica-Bold", size: min(videoRect.height / 10, 100))
            textLayer.shadowOpacity = 0.5
            textLayer.alignmentMode = .center
            textLayer.contentsScale = UIScreen.main.scale
            textLayer.isWrapped = true
            
            let textHeight: CGFloat = min(videoRect.height / 10, 120)
            let textWidth: CGFloat = videoRect.width
            let xPos = (videoRect.width - textWidth) / 2
            let yPos = (videoRect.height - textHeight) / 2
            textLayer.frame = CGRect(x: xPos, y: yPos, width: textWidth, height: textHeight)
            textLayer.zPosition = 1
            
            let parentLayer = CALayer()
            parentLayer.backgroundColor = UIColor.clear.cgColor
            parentLayer.frame = videoRect
            parentLayer.addSublayer(textLayer)
            
            let videoCompositionInstruction = AVMutableVideoCompositionInstruction()
            videoCompositionInstruction.timeRange = CMTimeRangeMake(start: lastTime - video.duration, duration: video.duration)
            let layerInstruction = AVMutableVideoCompositionLayerInstruction(assetTrack: videoTrack)
            layerInstruction.setTransform(videoTrack.preferredTransform, at: lastTime)
            videoCompositionInstruction.layerInstructions = [layerInstruction]
            instructions.append(videoCompositionInstruction)
            
            parentLayer.zPosition = 0
            mainParentLayer.addSublayer(parentLayer)
            
        } catch {
            print("Failed to insert track: \(error.localizedDescription)")
            return
        }
    }
}

mainParentLayer.frame = CGRect(x: 0, y: 0, width: maxVideoSize.width, height: maxVideoSize.height)
mainVideoLayer.frame = mainParentLayer.frame

mainComposition.renderSize = maxVideoSize
mainComposition.instructions = instructions
mainComposition.frameDuration = CMTime(value: 1, timescale: 30)
mainComposition.animationTool = AVVideoCompositionCoreAnimationTool(postProcessingAsVideoLayer: mainVideoLayer, in: mainParentLayer)

let outputUrl = NSURL.fileURL(withPath: NSTemporaryDirectory() + "merged" + ".mp4")

guard let exporter = AVAssetExportSession(asset: videoComposition, presetName: AVAssetExportPresetHighestQuality) else { return }

exporter.videoComposition = mainComposition
exporter.outputURL = outputUrl
exporter.outputFileType = .mp4
exporter.shouldOptimizeForNetworkUse = true

exporter.exportAsynchronously {

    DispatchQueue.main.async {
        
        if let outputUrl = exporter.outputURL {
            
            if exporter.status == .completed {
                
                self.play(video: AVAsset(url: exporter.outputURL!))
                completion(outputUrl, exporter)
                
            } else if let error = exporter.error {
                
                print("Export failed: \(error.localizedDescription)")
            } else {
                
                print("Export status:", exporter.status)
            }
        }
    }
}
//Originally played video here via AVPlayer, which played back immediately
//play(video: exporter.asset)
}

将视频与叠加文本合并

问题描述投票：0回答：1

1个回答

最新问题

将视频与叠加文本合并

问题描述 投票：0回答：1

1个回答

最新问题

问题描述投票：0回答：1