IOS - How to convert an UIImage to a CVPixelBufferRef with 420YpCbCr8BiPlanarFullRange format

Question

I'm trying to render frames from a video using OpenGL-ES with the specific format 420YpCbCr8BiPlanarFullRange.

The context is the following : I have processed a frame from my IOS device with OpenCV and then I convert it to an UIImage in order then to convert it to a CVImageBufferRef so that I could render it using OpenGL-ES.

So briefly : IOS device --> openCV Mat --> UIImage --> CVImageBufferRef --> Rendering using OpenGL ES

During the step of CGBitmapContextCreate I get this error :

Error>: CGBitmapContextCreate: invalid data bytes/row: should be at least 5184 for 8 integer bits/component, 3 components, kCGImageAlphaPremultipliedLast.
Jul 20 16:36:01
<Error>: CGContextDrawImage: invalid context 0x0. If you want to see the backtrace, please set CG_CONTEXT_SHOW_BACKTRACE environmental variable.

After checking all related post in Stack Overflow, I figure out that the problem of this error is because of the number byte per row. So I printed it out and I get this:

[INFO IMAGE] Width 1296.000000, Height 968.000000, Byte per row 1948

Indeed, according to this post, I don't have a number of bytes per row that is at least four times the width of the image represented.

But I really don't understand what is my error during the process. The UIImage is well represented and got 4 channels.

Here the code :

UIImage* imageTest = MatToUIImage(imageCV);
CGImageRef imageRef=[imageTest CGImage];
NSLog(@"Size UIIMage: (%f,%f)", imageTest.size.width,imageTest.size.height);

CVImageBufferRef pixelBuffer = [self pixelBufferFromCGImage:imageRef];

The main function which perform the conversion :

- (CVPixelBufferRef)pixelBufferFromCGImage:(CGImageRef)image
{
    CGSize frameSize = CGSizeMake(CGImageGetWidth(image), CGImageGetHeight(image));

    // set pixel buffer attributes so we get an iosurface
    NSDictionary *pixelBufferAttributes = [NSDictionary dictionaryWithObjectsAndKeys:
                                           [NSDictionary dictionary], kCVPixelBufferIOSurfacePropertiesKey,

    CVPixelBufferRef pixelBuffer = NULL;



   CVReturn status = CVPixelBufferCreate(kCFAllocatorDefault, frameSize.width, frameSize.height, kCVPixelFormatType_420YpCbCr8BiPlanarFullRange,
                      (__bridge CFDictionaryRef)pixelBufferAttributes, &pixelBuffer);

    if (status != kCVReturnSuccess) {
        return NULL;
    }
    NSLog(@"[INFO IMAGE] Width %f, Height %f, Byte per row %zi",frameSize.width,frameSize.height,CVPixelBufferGetBytesPerRow(pixelBuffer));


    CVPixelBufferLockBaseAddress(pixelBuffer, 0);
    void *data = CVPixelBufferGetBaseAddress(pixelBuffer);
    // get y plane
    const uint8_t* yDestPlane = reinterpret_cast<uint8_t*> (CVPixelBufferGetBaseAddressOfPlane(pixelBuffer, 0));

    // get cbCr plane
    const uint8_t* uvDestPlane = reinterpret_cast<uint8_t*> (CVPixelBufferGetBaseAddressOfPlane(pixelBuffer, 1));


    CGColorSpaceRef rgbColorSpace = CGColorSpaceCreateDeviceRGB();
    CGContextRef context = CGBitmapContextCreate(data, frameSize.width, frameSize.height,
                                                 8, CVPixelBufferGetBytesPerRow(pixelBuffer), rgbColorSpace,
                                                  kCGImageAlphaPremultipliedLast | kCGBitmapByteOrder32Little);

    CGContextDrawImage(context, CGRectMake(0, 0, CGImageGetWidth(image),
                                           CGImageGetHeight(image)), image);
    CGColorSpaceRelease(rgbColorSpace);
    CGContextRelease(context);
    CVPixelBufferUnlockBaseAddress(pixelBuffer, 0);

    NSLog(@"[INFO IMAGE] return pixel buffer done");
    return pixelBuffer;
}

Here an output from my Pixel Buffer :

<CVPixelBuffer 0x127f5bf60 width=1296 height=968 pixelFormat=420f iosurface=0x12b900628 planes=2>
<Plane 0 width=1296 height=968 bytesPerRow=1296>
<Plane 1 width=648 height=484 bytesPerRow=1296>
<attributes=<CFBasicHash 0x127f47260 [0x1a1462150]>{type = immutable dict, count = 2,
entries =>
    0 : <CFString 0x19ce8fca8 [0x1a1462150]>{contents = "PixelFormatDescription"} = <CFBasicHash 0x127e7f900 [0x1a1462150]>{type = immutable dict, count = 10,
entries =>
    0 : <CFString 0x19ce8ff08 [0x1a1462150]>{contents = "Planes"} = <CFArray 0x127e7f6f0 [0x1a1462150]>{type = mutable-small, count = 2, values = (
    0 : <CFBasicHash 0x127e7f940 [0x1a1462150]>{type = mutable dict, count = 3,
entries =>
    0 : <CFString 0x19ce901a8 [0x1a1462150]>{contents = "FillExtendedPixelsCallback"} = <CFData 0x127e7fac0 [0x1a1462150]>{length = 24, capacity = 24, bytes = 0x000000000000000030535685010000000000000000000000}
    1 : <CFString 0x19ce8ffa8 [0x1a1462150]>{contents = "BitsPerBlock"} = <CFNumber 0xb000000000000082 [0x1a1462150]>{value = +8, type = kCFNumberSInt32Type}
    2 : <CFString 0x19ce8ffc8 [0x1a1462150]>{contents = "BlackBlock"} = <CFData 0x127e7f980 [0x1a1462150]>{length = 1, capacity = 1, bytes = 0x00}
}

    1 : <CFBasicHash 0x127e7fb50 [0x1a1462150]>{type = mutable dict, count = 5,
entries =>
    2 : <CFString 0x19ce8ffe8 [0x1a1462150]>{contents = "HorizontalSubsampling"} = <CFNumber 0xb000000000000022 [0x1a1462150]>{value = +2, type = kCFNumberSInt32Type}
    3 : <CFString 0x19ce8ffc8 [0x1a1462150]>{contents = "BlackBlock"} = <CFData 0x127e7fbb0 [0x1a1462150]>{length = 2, capacity = 2, bytes = 0x8080}
    4 : <CFString 0x19ce8ffa8 [0x1a1462150]>{contents = "BitsPerBlock"} = <CFNumber 0xb000000000000102 [0x1a1462150]>{value = +16, type = kCFNumberSInt32Type}
    5 : <CFString 0x19ce90008 [0x1a1462150]>{contents = "VerticalSubsampling"} = <CFNumber 0xb000000000000022 [0x1a1462150]>{value = +2, type = kCFNumberSInt32Type}
    6 : <CFString 0x19ce901a8 [0x1a1462150]>{contents = "FillExtendedPixelsCallback"} = <CFData 0x127e7fc80 [0x1a1462150]>{length = 24, capacity = 24, bytes = 0x0000000000000000ac515685010000000000000000000000}
}

I checked this post, this one and others but did not find something which works in my case.

Could you please help me ? I checked all the related posts and I did not find a way to make it work properly.