10% off monthly plans with code NODE10, until August 31

Imaginode
All tools

Fashion transition: the outfit that melts into liquid

The outfit melts from top to bottom into glossy liquid, and the new one solidifies in its place. The transition of fashion reels, computed between a first and a last frame. Full demo visible without an account.

Interactive demo

Real workflow, real results. Pan and zoom freely.

Use this workflow

Both states are there: the imported photo (outfit A), the Flux Kontext render (outfit B), and the video melting from one to the other, first and last frame wired. Open the nodes to read the instructions.

At a glance

What the workflow produces
1 image in 9:16 · 1 video of 5s in 1080p, 9:16
Canvas structure
6 nodes wired by 5 connections
Models wired in
Flux Kontext, Kling V3
Cost of one full run
about 158 credits, roughly €1.58

The melting effect is the fashion transition of the moment: the subject melts like colored wax, the clothes stream down as glossy liquid, then the liquid rises back and solidifies into the new outfit. On a canvas, the effect needs no frame-by-frame retouching: one photo, one edit, one cable.

The mechanic is the one behind the before/after, applied to clothing: the imported photo becomes the FIRST frame of the video, the edited render the LAST, and the model computes the melt between the two. The edit is the precision point of the workflow: Flux Kontext changes only the outfit, same person, same pose, same background. If the pose shifts between the two frames, the transition jumps instead of melting.

The video runs on Kling V3, five seconds in 9:16 with generated audio, you can hear the dripping. The guide consistent character explains why fidelity to the subject is decided in the model inputs, and the calculator gives the clip price before creating an account.

How it works

  1. 1

    Import the full-length photo

    Outfit A, framed full-length, plain background, simple pose facing the camera. The melt needs the whole body and a bit of floor: the liquid must have somewhere to pool.

  2. 2

    Describe outfit B, and nothing else

    The edit instruction replaces the clothes and locks everything else: same person, same face, same pose, same background, same light. That is the rule that holds the transition together.

  3. 3

    Check both frames

    The photo is wired as first frame, the render as last. If the render moved, shifted pose or changed background, rerun the edit before paying for the video: the melt cannot fix a disagreement between the two states.

  4. 4

    Generate the melt

    The video prompt describes the material movement: melting from top to bottom, liquid streaming and pooling, then rising back and solidifying. Five seconds, static camera, audio on.

Variations and use cases

The same canvas, tuned for different needs: each variation is two or three lines of prompt away.

Before/after makeover

The jeans-and-sneakers melt into the tailored suit: the transition tells the makeover better than a hard cut. Hairdressers, stylists and image coaches get their signature format here, from a single client photo. The edit instruction only touches the clothes, the face stays exactly the same.

Collection launch

Last season's piece melts into the new one: same model, same pose, only the collection changes. Declined across three silhouettes, the effect makes a complete launch campaign. The AI photoshoot builds the starting poses if the collection has not been shot yet.

Team jersey and merch

The home jersey melts into the away one, the tour tee into the new date: the most effective merch announcement is the one where the old product becomes the new one before your eyes. Keep the background neutral: the liquid must stay the only thing moving.

Period costume and cosplay

Streetwear melts into armor, a modern dress into a 1900 gown: the gap between the two outfits is the value of the clip. The further apart the two states, the more the melt impresses, as long as the person and the pose do not move a pixel.

The first/last frame mechanic transforms any subject: a living room melting into its renovated version instead of the classic home staging, a package into its new design, a logo into its rebrand. Only the element to transform may change between the two images.

Common mistakes

Changing the pose along with the outfit

If the edited render moved an arm or turned the head, the model can no longer melt: it cuts or distorts. The edit instruction must lock the person, the pose, the background and the light, and only free the clothes.

Describing the melt in the edit instruction

Each prompt has its role: the edit builds the clean final state, the video tells the melt. An end image already melting or dripping robs the video model of its target, and the transition ends in a puddle instead of solidifying.

Framing too tight

The melt needs the whole body and a bit of floor: the liquid streams downward and must have somewhere to pool before rising back. A bust framing cuts the movement mid-course and the effect loses its logic.

Frequently asked questions

Why go through an edit rather than two photos?

Because the transition demands the same person in the same pose to the pixel, and two real shots never guarantee that. The Flux Kontext edit only touches the clothes: everything else is identical by construction.

Can I still use two real photos?

Yes: replace the Image node with a second Media node and wire it as the end frame. Take both photos without moving the camera or the subject, ideally on a tripod: the slightest framing gap shows up in the melt.

Does the effect work on things other than clothes?

Anywhere two states of the same subject exist: a package melting into its new version, a room melting into its renovation, a logo into its rebrand. The rule stays the same: only the element to transform changes between the two frames.

Why does my transition jump instead of melting?

Almost always because the two frames disagree: shifted pose, changed background, different framing. Check the edited render before generating the video, and rerun the edit until only the clothes differ from the photo.

Which video models accept an end image?

Kling V3, the template's choice, but also Seedance 2.5 and Veo 3.1 Fast. The canvas only shows the « end image » cable on models that support it.

How much does a full transition cost?

The Flux Kontext edit costs a few cents, the five-second video around a dollar: count about $1.30 for the finished clip, sound included. The calculator details the exact workflow cost before creating an account.

Can I chain several outfits in one reel?

Yes, in cascade: the outfit B render becomes the first frame of the next transition, towards an outfit C. Each segment stays a five-second clip, the edit chains them, and the person never moves between two melts.

Does the transition work from a phone photo?

Yes, that is the typical use: a full-length photo, sharp, simple background, even light. Avoid busy backgrounds, crowds or backlight: anything visually moving between the two frames starts shaking during the melt.

Fashion transition: the outfit that melts into liquid

Use this workflow

Free account, no card required. The workflow opens pre-filled in your canvas.