Almost all photographers use some amount of post-processing. Editing and manipulating photos, after the photo has been taken, has always been a part of the photographic process. Ansel Adams was famous for modifying and manipulating his photographic prints. There is no such thing as a “raw” or “unchanged” or “unmodified” photograph. Even basic digital cameras have presets that will customize the look of the image based on the settings within that camera. Not editing a photo is just using the edit that the camera maker provided. (Which in many instances is just fine.)
The important question for any photographer isn’t, “should I edit my photos?” It is, “when does editing become a problem and how much editing am I willing to use?” Based on the above statements, I clearly endorse changing the contrast, saturation, and brightness of an image. But is changing a blue sky to a sunset a manipulation of contrast and saturation? While technically the answer is yes, I would never do this. I draw the line for editing at “changing the character or key elements of a scene.”
In the example above, a scene with a blue sky was taken during the day. Changing that sky to a sunset has changed key elements. That scene no longer truthfully shows the correct time or weather in that scene. I think this is a fairly obvious example of my philosophy on editing. A trickier example is removing a person or object. If that person or object is “important” than it should remain, no matter how unsightly. It is probably easier to describe “important” by describing what is “unimportant.” A small blurry figure on the edge of a frame is not important. A powerline that cuts across the top corner of the sky is not important. A single small person that stuck their head out from behind a wall is not important. These elements do not change the character in any way.
Another simple criterion for acceptability of editing an image can be the physical size of the change. Generative AI makes large changes much easier than they used to be. If there is a bench in an image that I want to remove, it used to be hard to do this if the bench is big in the frame. Removing the bench would leave a big hole in its place. This could be fixed by copying nearby areas of the image, but this requires skill and gets harder as the area gets larger. I don’t really endorse this practice, but the difficulty of copying large areas at least made it a challenge. Generative AI can now just invent something to fill gaps. If I want to remove something, but need AI to invent a whole new image to replace that something, it should not be removed in the first place. In this instance, a picture of something deemed ugly has been replaced by something prettier that a computer invented to replace it. This is not art.
This website uses cookies to improve your experience.