I’m a complete newbie to Photo AI. Am using Lightroom Classic and Photo AI as an add-in. My camera is a Lumix DMC-FZ300.
I selected several RW2 images in LrC and copied them to Photo AI. All photos were processed on “Autopilot”. In addition, I turned on Sharpen. Everything else was left in the default state.
One of the images is of a rectangular sign. Photo Ai changed all four of the mostly straight edges so that they are now curves.
It also increased the width of the exposed wood edges behind the sign. I think this is because the RAW mode on my camera captures additional pixels at the edges.
The screenshot is from LrC.
Is this a bug or do I need to change some of the settings?
Thanks for reaching out. This is due to Topaz Photo AI not having lens correction for this Raw file. The lens has heavy distortion that we are not able to correct.
A workaround for this is to use the Photo > Edit in > Topaz Photo AI method instead of the Raw workflow which will preserve the lens correction when converting the image to a TIF.
Could you send me the Raw file so I can check the lens correction with my team? You can securely submit your image(s) to my Dropbox using the link below. Please be sure to send me a note to let me know you sent something. Dropbox File Request
Hi Lingyu,
Thank you for your response. I have uploaded 6 files to your Dropbox. Each set consists of an RW2, DNG, and XMP file. There is one set corresponding to this post (P1200306).
The other set (P1200293) relates to another issue that I posted earlier today (Photo AI is vignetting my RAW (or ignoring the lens profile)), wherein some of my images come back from Photo AI vignetted. In that post, I speculated that the problem was related to lens correction, and based on your response here, I think I was right.
Also in that post, I noted that the lens correction information is apparently embedded in the RW2 file so that it can be used by Lightroom. Could Photo AI not do the same, utilize the embedded lens correction information?
I saw the other thread and yes it is the same issue. The distortion will change based on the lens and the zoom used with that lens.
The information may be in the file, but unfortunately there is no instruction on how to apply that information to correct the lens. Panasonic does not have public documentation on how those numbers can be translated into a corrected image.
We have reached out to them to get more information and hopefully be able to support those files. For now, it would be best to use the Photo > Edit in > Topaz Photo AI workflow.