我要求在实时视频上应用过滤器,我正在尝试在Metal中进行过滤. 但是在将过滤器编码为目标过滤器之后,我遇到了将MTLTexture转换为CVPixelBuffer的问题.参考(https://github.com/oklyc/MetalCameraSampl
但是在将过滤器编码为目标过滤器之后,我遇到了将MTLTexture转换为CVPixelBuffer的问题.参考(https://github.com/oklyc/MetalCameraSample-master-2)
这是我的代码.
if let pixelBuffer = pixelBuffer { CVPixelBufferLockBaseAddress(pixelBuffer, CVPixelBufferLockFlags.init(rawValue: 0)) let region = MTLRegionMake2D(0, 0, Int(currentDrawable.layer.drawableSize.width), Int(currentDrawable.layer.drawableSize.height)) let bytesPerPixel = 4; let bytesPerRow = CGFloat(bytesPerPixel) * currentDrawable.layer.drawableSize.width let tempBuffer = CVPixelBufferGetBaseAddress(pixelBuffer) destinationTexture.getBytes(tempBuffer!, bytesPerRow: Int(bytesPerRow), from: region1, mipmapLevel: 0) let image = self.imageFromCVPixelBuffer(buffer: pixelBuffer) CVPixelBufferUnlockBaseAddress(pixelBuffer, CVPixelBufferLockFlags.init(rawValue: 0)) }
imageFromCVPixelBuffer方法看起来像这样.
func imageFromCVPixelBuffer(buffer: CVPixelBuffer) -> UIImage { let ciimage = CIImage(cvPixelBuffer: buffer) let context = CIContext(options: nil) let cgimgage = context.createCGImage(ciimage, from: CGRect(x: 0, y: 0, width: CVPixelBufferGetWidth(buffer), height: CVPixelBufferGetHeight(buffer))) let uiimage = UIImage(cgImage: cgimgage!) return uiimage }
这是通过金属渲染图像的屏幕截图
以下是将MTLTexture转换为CVPixelBuffer的同一图像的屏幕截图.
将MTLtexture转换为CVPixelBuffer需要写入AVAssetWriter,然后将其保存到库中.
不要那样自己计算bytesPerRow.它被传递给Metal以让Metal知道如何排列行.您希望Metal以CVPixelBuffer期望的方式排列它们.因此,您应该使用CVPixelBufferGetBytesPerRow()来确定值.