Hacker News Reader

Comment by jcynix | Hacker News Reader

You can transfer EXIF information, including the rotation info, with either exiftool or ImageMagick into a thumbnail, to let other software know how to display it correctly.

If you want to rotate a jpeg yourself (which is easily done with "magic in.jpg rotate 90 out.jpg" and, of course, other software) but lossless, there's https://jpegclub.org/jpegtran/ which accomplishes it.

When generating rotated thumbnails (without caring for "lossless") the best strategy in my experience is to first rotate the large original image and then generate the thumbnail afterwards, because image resizing will most often hide any rotation artifacts anyways.

View on HN

But it seems that if you are re-encoding the image anyways it is best to just generate it in the "correct" orientation and skip this metadata. Emitting the metadata seems mostly useful if you are trying to avoid re-encoding the image.

Or is there a benefit to generating the thumbnail in the original rotation then preserving the metadata?

Sure, generating the "correct" format would be a good decision. When you change pixel orientation, you'll need to either update or clear the EXIF orientation info to reflect the new pixel data.

Rotation metadata might be helpful in (larger) thumbnails when images are displayed on mobile devices and responsive web pages. Metadata is easy to transfer, at least when scripting with Exiftool or ImageMagick.

Is jpeg the only format which can be “rotated losslessly” in this way, ie without exif metadata (and excluding any obvious formats like bitmaps or whatever can be trivially rotated) ?

Both PNG and Webp exist in lossless formats which can be easily rotated without loss of quality. Jpeg is different, because its internal structure consists of 8x8 blocks (which, BTW, can be rotated lossless if both sides of the image are multiples of 8, iirc.

For photos all these assumptions are rather theoretical IMO, because modern cameras create images in sizes which almost always have to be reduced in size for viewing, unless one wants to pixel-peep.

From what I know it can't be "rotated losslessly" in all cases, only if the dimensions of the images are multiples of the MCU which are block of pixels whose size is determined by the chroma subsampling. Ex.: With the common "4:2:0" subsampling the MCU is 16x16 the image's height and width must be exactly multiples of 16, otherwise I think it's just visually lossless and uses some tricks that I'm still not sure how they work.

Reply on news.ycombinator.com