Resolution and DPI get used interchangeably in tutorials and product pages, which is confusing because they measure different things. This explains what each one actually means and how the distinction matters when you are preparing images for print or web.
An image is a grid of pixels. Resolution is the number of pixels horizontally and vertically. A 1920 by 1080 image has 1920 columns and 1080 rows of pixels.
Resolution is the only thing that actually exists in the file. Everything else is computed from it.
Higher resolution means more detail is available. It does not mean higher quality, because quality also depends on the source and on how the image was captured or generated.
DPI stands for dots per inch. It is only meaningful when an image is being printed, because it describes how many pixels the printer should place in each physical inch.
A file does not have an inherent DPI. It has a resolution and, optionally, a metadata tag that says "when you print this, treat 300 pixels as one inch." Software reads that tag and calculates the physical print size.
This is why the same file can print at 4 by 6 inches at 300 DPI or at 8 by 12 inches at 150 DPI. The file did not change. Only the interpretation changed.
Background removal is a pixel level operation. It runs on the resolution of the file, not on any physical size.
If you process a 500 pixel image and then try to print it at 4 by 6 inches, the print will be low quality because 500 pixels spread over 4 inches is only 125 DPI. The background removal did not cause this. The source resolution did.
The practical rule: process at the highest resolution you have, then resize for the output medium. Do not process a small file and try to upscale it afterward. The AI mask is only as detailed as the input.
Social media: 1080 pixels on the short side is the practical minimum. Instagram and TikTok both downscale to this.
Web images: 1200 to 1600 pixels on the longest side is enough for full screen display on retina monitors.
Ecommerce marketplaces: 2000 pixels on the longest side enables the zoom feature on Amazon and most other platforms.
Print at 4 by 6 inches: 1200 by 1800 pixels for 300 DPI.
Print at 8 by 10 inches: 2400 by 3000 pixels for 300 DPI.
Print at A4: 2480 by 3508 pixels for 300 DPI.
Upscaling adds pixels that were not in the original. Interpolation guesses what the new pixels should be based on neighbors. It cannot recover detail that was never captured.
For an AI generated image, upscaling can look acceptable because the original already had smooth gradients. For a photo, upscaling almost always produces a soft, muddy result that is worse than the original at native size.
If a client asks for a higher resolution than you have, the honest answer is to reshoot at higher resolution or to accept the lower print size. Do not promise an upscale that will not deliver.
No. The DPI tag is a few bytes of metadata. The pixels are unchanged. Changing the DPI tag changes only the physical size at which the image prints by default.
PPI is pixels per inch and is used for screens. DPI is dots per inch and is used for printers. In practice people use them interchangeably. The technical difference does not matter for most workflows.
Zooming is the same as asking the software to interpolate. At 100 percent zoom you see native pixels. Above that you are seeing the software guess at missing pixels. This is not a quality problem, it is a resolution limit.
Yes, but it only affects print interpretation. If you have a 1000 pixel image and set DPI to 300, the software will print it at 3.33 inches wide. If you set DPI to 100, it prints at 10 inches wide with visible pixelation.