Real-ESRGAN GUI, a graphic interface of Real-ESRGAN an image magnification tool.
https://sourceforge.net/projects/real-es...ui.mirror/
https://github.com/bycloudai/Real-ESRGAN-Windows
A test :
original :
   
upscaled
   

The attached screenshot shows an image I am attempting to use as a logo for a website. However, whenever I add the image the banner increases in height. After playing around with the code and not being able to figure it out, I can only assume it's because of the background size. I've already scaled the image as large as I can. Is it possible to crop the red portion of the screenshot, so all I see is the image and none of the background?

Attached Files

Thumbnail(s)
   

Unfortunately, the documentation I've been able to find says little to nothing about what is actually being done to the image when the "Auto Input Levels" button is selected on the "Color>Levels" tool.  The GIMP User Manual section says only "Auto: Performs an automatic setting of the levels." which is applied to All Channels.

This question arises as a result of what I'd call quite remarkable results that are produced by the Auto Input Levels when NOT being able to come even close to getting the same results by using the other available (e.g. manual) options for making adjustments.

The related image files were produced by scanning some pretty old film strips (35mm negatives).  For example, when the scanner produced image file is opened in GIMP and a jpg image is created before making any edits the result is as follows:




Then when doing nothing other than applying "Color>Levels>Auto Input Levels" the similar jpg image file that results follows:



To the extent that it might matter a screenshot of the tool ("Adjust Color Levels") values associated with the original image file shown above is as follows:



Note:  When the "Auto Input Levels" button is selected the image is changed quite dramatically in this case but nothing gets changed/updated within the "Adjust Color Levels" window.

These examples are derived from using GIMP 2.20.36 on Windows.

My thinking is that "Auto" means the tool is doing the same things described in the documentation that I should be able to do manually.  One might even think that the window which shows changes when made manually could be updated to reveal what changes that are being made automatically.  Of course this is NOT possible if "Auto Input Levels" is doing something different than what a user can do on there own.  If so, it would be nice to at least know what kind of things are possible when selecting "Auto Input Levels" that cannot be done otherwise.

The lefthand menu pane of my software vanished while I was busy working with GIMP, so clearly I pushed something I should not have pushed. I've tried: 

• Pushing tab
  
• Cnt B
  
• Looking for an option in the "View" menu
  
• Looking for an option in the "windows" menu. 
  

Suppose I have a pure black (#000000) layer and a pure white (#FFFFFF) layer.
They are one on top of the other, both with normal blending. The one above is 50% opacity.

I'd expect that the resulting color would be a midpoint gray (#7F7F7F). And I'm pretty sure that's what I used to get on GIMP until recently, and it's still what I get on other softwares.
Instead, I'm getting #BCBCBC - a far brighter color.

This is not just about layer blending either - any brush with some kind of transparency or anything like that - the result of mixing colors always come up brighter than I'd expect.

I tried both "Perceptual Gamma" and "Linear Light" modes, no change.

Video demonstrating (I compare with Paint.Net on it, but the result there is the same on any other software aside from GIMP - from Photoshop to html rendering in web browsers):
https://www.youtube.com/watch?v=OfAXikmC...e=youtu.be

I was wondering what happens to metadata in Gimp. I was having fun playing with some programming languages to create an image editor. I understand that in broad terms the structure is this:

I have a .jpg photo that I want to edit, I upload it to the editor, the editor generally uses an external library to read the .jpg file; it takes this file, decompresses it, isolates the bits relating to the actual image which will form an array of pixels of the exact dimensions of the image that will be shown on the screen.

At this point each editor will have its own functionality for editing the photo.

Once you have achieved the desired result you can export the photo again as .jpg. The edited pixel array is taken and compressed again into a .jpg file via an external library.

I wanted to understand what happens to the metadata in this process in Gimp. I tried to read the Gimp source code but I didn't understand.

I'm interested in understanding well what the final exported .jpg image is made of. If it is composed of the compressed pixel array of the edited image + some metadata created by Gimp useful for the final file (Such as resolution and size) + any metadata from the original file that Gimp recognizes and copies into the new one (Asking you during export if you want to keep them or not and allowing you to modify them).

Or could it happen that Gimp adds metadata present in the original photo to the final .jpg but does not recognize it and does not show it in the metadata viewer or editor?

If anyone knows any places within the Gimp code where I can understand this, it would be very useful for me.

I have a background layer and I want to paste a transparent image to it. I.E I do a Google image search for ‘trees transparent’ and receive a full page of results. However when I select a result with a checkered background vs an image with a white background. Only the ones with the white background copy and paste correctly. The ones with the initial visible checkered background do not. (Pasted image also includes the background with these). I was wondering why..

Sorry, I'm a beginner in gimp. 
I have a picture with white text on black background. First I want to remove the text. I try using the klone tool. I use the ctl-key for marking the source spot. But the klone tool does not do anything. 
I would appreciate a help tip

Hello, would love to see feature like blender or gimel the node-based editing

I'm new here, so hi.

After optimising for an animation, is there a way of exporting the list showing in the Layers Editor as a text fiile?

For Example:
----------------

img001.jpg(100ms)
img002.jpg(2000ms)
img003.jpg(200ms)
.... etc

cheers Sean.