I was an avid user of Autopano by Kolor and then GoPro bought them and decided to just shut down the company. Not sure why but dead it is. PTGUI is the only real other player in town and the Pro version, the only one that can handle HDR, is over $300USD.
I’m sure DXO’s expertise in lens corrections and such could build out a panorama stitcher that could rival all the other ones out there. A stitcher for gigapixel images that won’t bog down a system. It would be beyond awesome if they came up with a way to host these images on a website too. Something like a Wordpress plugin would be great to see.
That is also what I do with Autopano Pro. I develop my Raw files in DXO - just optical corrections - and the rest in Autopano. I still have to try out Panos with Deep Prime before sending them to Autopano, never tried that sofar.
I don’t believe any program stitches RAW files. Lightroom makes you believe it does just that but it produces a non-RAW DNG file which indicates that files are demosaiced first. I can’t see how you would stitch un-mosaiced files.
I process my RAW files in PL using just lens correction then stitch in one of a number of panorama programs (some do better jobs on different photos).
That was the beauty of AutoPano was it had RAW conversion built it and exporting also included 32-bit EXR files which practically kept the content as close to RAW as you possibly can. Sure it created absolutely monstrous files but what you could do with them, already stitched in PShop was awesome.
I’ve watched the progression of DXO over the years and it looks like they are in a specific niche that protects them in a way while now also encroaching on Adobe’s turf.
High-quality, functional pano software seems to be one niche that hasn’t had much attention over the last few years. Those two points are why I suggested DXO look into creating ‘The Ultimate’ pano stitching suite at an affordable price. I’d be in in a hot minute.
Can I stitch RAW images? Yes, but it’s not “recommended”. Autopano is stitching software, not RAW decoding software, so Autopano will only decode the RAW files with the very default values, defeating the whole purpose of RAW shooting. It is best to treat your RAWs in dedicated software beforehand. Prefer shooting in JPG or TIFF if you do not want to treat the RAW photos.
Autopano is able to decode the RAW format. However, the different RAW formats evolve over time, depending on brands and camera models. Therefore, it is possible that Autopano can’t decode some RAW files, simply because the database is not updated. This usually happens when a new camera model is released, containing a new coding of its RAW format and which is not included in Autopano. The solution for that is to decode this new RAW format in .tiff 16 bits format, which is the equivalent of what Autopano does during the decoding. We update our decoding database following this evolution, but it takes some time, so there is a delay between the release of a camera and the update of a new version of Autopano that support this new encoding format.
As you can see there’s a lot of arguments to have the pano stitching function not to be part of a raw converter.
Maybe sending a number of files to a pano stitching program would be some solution. But it won’t change anything. The files are saved to disk and a program is called with the files as a parameter.
I am also user of Autopano ; it is still usable but with somme crashes.
It has been killed by GoPro and I am waiting too for this function.
I believe that code libraries for stitching are common to all softwares and free, it is a problem of interface, so I hope…
I would also vote for adding panorama stitching to DxO PhotoLab. My present workflow (shooting composite images of murals) is an awkward blend of software that involves GIMP, DxO ViewPoint, Kolor Auto Pano Giga, and DxO PhotoLab. Integration of panorama stitching into PhotoLab would make life a whole lot easier.
I’ve found that panoramas shot from a single vantage point with a handheld camera can be sewn together nicely with either Auto Pano Giga or PTGUI. I presume that this minimal functionality would be easy for DxO to replicate. When the overlap of images is poor, I can often construct a decent panorama by applying a perspective correction first and then sewing with APG. PTGUI is very unhappy with this approach, apparently because it really expects to find a common camera position for all the images it sews together, and the perspective corrections confuse it greatly.
A related issue, where APG is able to produce composite images and PTGUI is not, is when the individual photos are shot from different positions. Sometimes this is necessary, as when there is glare on the subject or when there is an obstacle in front of some part of the subject. APG can often construct a composite, although post-processing is necessary to correct some overall distortions.
I’m not sure what APG does algorithmically that PTGUI cannot do, but when DxO develops panorama sewing capabilities, I would hope they can match the remarkable flexibility that APG offers.