perdón.

Step by step4 min read

How to make the Perdón Trend from one original photo

Start with one photo, review three AI edits and keep the original last. Add one confession per frame, download the MP4 and finish the music in TikTok.

Before you start

  • Plan the ending first: the original photo belongs on the last slide.
  • Remove one change per image so that each swipe tells part of the story.
  • Download a captioned MP4 or four vertical PNGs, then add music in TikTok.
  • Captions, preview and export are free. Generating the AI edits requires an account and credits.

1. Choose a photo that works as the ending

The Perdón Trend opens with the most edited image, but you build it from the other end: an original photo that can deliver the final reveal. An everyday selfie, a pet on the sofa or a simple workout photo can all work. The story is about discovering what the earlier versions added.

Pick a clear image with a visible subject and framing that works on a portrait canvas. A tiny face, a hand covering the eyes or a background full of reflections makes the differences harder to compare. You do not need a studio portrait. You need a reference that will remain recognizable across the sequence.

Keep a copy of the source and use photos you own or have permission to edit. Reserve that source for the final frame. Perdón Studio fits the whole original to the canvas and adds your caption, without cropping it or applying another AI transformation.

2. Choose a template and review the story

Templates include the three planned edits and starter captions, so you do not need to write four prompts. Choose a story that fits your photo and review its plan. A reveal is easier to follow when the viewer can point to what changed: start with the most transformed version and remove one edit at a time until the original appears.

Here is a proposed portrait storyboard. Adapt it to what is actually visible in your original. This is a planning example, not a set of generated results.

  1. 01

    The full edit

    An edited jacket, hairstyle and background.

    This is me.
  2. 02

    First confession

    The original background returns; two edits remain.

    That place wasn’t real either.
  3. 03

    Second confession

    The original clothing returns; one edit remains.

    The jacket was AI too.
  4. 04

    The original

    The starting photograph is revealed.

    Sorry. This is actually me.

3. Prepare three edits with controlled differences

In an AI image editor, supply the same reference photo with each request. Describe both what should change and what should stay: identity, expression, pose, framing and light, unless one of them is deliberately part of the story. Asking for a better-looking version leaves too many decisions open.

Work from the original reference when your tool supports it. Repeatedly editing a generated image can accumulate changes you did not request. Compare the eyes, hands, silhouette and background lines after each attempt. No prompt guarantees that an image model will preserve every detail.

A Perdón Studio template needs just one original photo from you. The studio checks AI availability before generation. When it is available, sign in, review the credit cost and create the three versions for the chosen story. You do not need to upload four photos. Check each result before downloading: an edit can change details you meant to preserve.

4. Arrange the images and check your captions

Place the most edited version first, followed by the two partial reveals and then the original. Review the whole sequence and each adjacent pair. The differences should be visible without the face or camera seeming to jump into a new scene.

Use a short sentence on each slide. A specific confession such as the jacket was AI too gives the viewer more to notice than repeating sorry on every image. Keep text away from the face and leave some space around the edges, where the posting app may place its controls.

  • Can you identify one clear change between every pair of images?
  • Does the last frame show the original photo you chose?
  • Does each caption describe the edit that was just revealed?
  • Is the text readable when you view it at phone size?

5. Download the MP4 or photos and post on TikTok

Once all four images are ready, the MP4 button creates a 720 × 1280 portrait video with 3.5 seconds per image, 14 seconds in total. It contains four still images with captions and fixed timing; exporting does not generate movement between them. The file is created in your browser without using more credits. You can also download four 1080 × 1920 PNGs in a ZIP or an individual image.

Upload the MP4 to TikTok for a sequence with fixed timing. If you want viewers to swipe through photos, create a photo post with the PNGs in the same order. Add music within TikTok and review the options available to your account. Perdón Studio does not bundle music or post for you. If your browser cannot export MP4, the editor explains this and the PNG downloads remain available.

Before publishing, check the caption, audience and disclosure for AI-created content. TikTok says AI content containing realistic images must be labeled. Revealing the original at the end does not, by itself, replace that context.

Put the story together

One original. Your photo reveal.

Start with one photo and a template, or use existing edits in manual mode. Review the three edits, keep the original last and download a captioned MP4 or PNGs. Export is free; AI generation uses credits.

Open Perdón Studio