Cineground AI workflows

AI workflows · Refocus

Refocusing a shot with AI: first and last frame method

How we refocused a 10-second shot, from the house onto the car and the man, keeping the action and the grade: the method, what held up, and what did not.

By Arno Faure, Cineground, Montréal11 November 2026Draft 0.1About 6 minutes to read
Download the PDFContact
10 stest shot, “Arrival at Dusk”
2frames refocused, the first and the last
2video models compared, Seedance 2.5 and Kling 4

Flux 3 image edit

Frames
2
Output
2048 × 1152
Prompt expansion
off

Seedance 2.5 image to video

Output
1920 × 1080
Length
10.0 s

Kling 4 image to video

Output
1920 × 1080
Length
10.0 s

We were asked to refocus a shot. In the original, the house in the background is sharp, while the car and the man in the foreground are soft. We wanted the opposite, with the action and the framing unchanged: the car and the man sharp, the house and the trees softly out of focus. Today’s video models can change what is in a shot very easily (a car can become a pink unicorn car or a duck car), but refocusing is not an edit they offer. We understand a model for it is being trained, but it is not available in a simple form yet. Our method is to refocus the first and the last frame with an image model, then have a video model recreate the shot between them. The result is close to the original action, but it is a re-creation, not an edit, and it is not identical.

Key points

1The brief, and why it is not a standard edit

Video models that accept an input video (edit modes) understand instructions about content: replace an object, change a colour, change the weather. They do not understand an instruction about the lens. A refocus changes which parts of the frame are sharp, which is closer to a depth-dependent blur that must follow the moving subjects.

Type of changeStatus todayComment
Replace an object (car to unicorn car or duck car)EasySupported by edit modes, works on the first try.
Refocus, keep everything elseNo direct toolA dedicated model is reportedly being trained, not available in a simple form.
Recreate the shot from refocused first and last framesWorks, approximateThe method of this note. The action stays close to the original but is not identical.

Two examples on the same shot, each against the original. Move over the picture to compare.

Original
0:00 / 0:00
Original
Original
Pink unicorn car
Pink unicorn car
Duck on wheels
Duck on wheels

2Method

01First and last frameTaken from the original shot
02Refocus with Flux 3Image edit, car and man sharp, house and trees soft
03Character referenceFull-body image, character sheet, 360° turnaround video
04Time-coded promptRead from the original shot, second by second

One video model per version

Seedance 2.5Start and end image, turnaround video, prompt.
Kling 4Start and end image, reference images, prompt.
05CompareAgainst the original, by eye, on video and on frames

2.1 Material

ItemFormatUse
Original shot1920 × 1080, 10.1 s, with soundReference for the action and the framing
First and last framePNG, 1920 × 1080Input of the image edit
Refocused first and last framePNG, 2048 × 1152Start and end images of the video model
Character sheet and 360° turnaroundH.264, 1920 × 1080, 10 sIdentity reference for the video model
Generated versions (two engines)1920 × 1080, 10 s, with soundResults to compare with the original

2.2 Refocusing the frames with Flux 3

The first and the last frame of the shot are taken from the original and passed to Flux 3 as an edit. The instruction names what stays sharp (the car, the man) and what falls out of focus (the house and the trees), and says that composition, light and colour do not change. Flux 3 follows the frame closely and gives clean textures. The faces are the weak point: they stay close to the source but are not the same.

Original Refocus
OriginalRefocus
Original
Original
Refocus
Refocus
Why only two frames. Each Flux generation is unique. Two frames from the same shot never get exactly the same blur or the same micro-detail. At the start and the end of one shot this is invisible. Across a whole shot, or between neighbouring shots, it would show.

3Keeping the actor

The man is small in the frame and out of focus in the original, so the frames give no clean reference for his face. The video model then invents a face, and it changes during the shot. We built a reference outside the shot:

  1. Take a full-body image of the character from another shot of the same film.
  2. In Flora, generate a character sheet from it, then a 360° turnaround video, with the Character 360 kit technique.
  3. Use only the turnaround video as a reference in the video prompt, next to the start and the end image.
360° turnaround
0:00 / 0:00
Split-frame turntable
360° turnaround

4Prompting

The prompt states what each reference is, fixes the camera and the vehicle, and describes the action second by second. The action must be described precisely, because the model recreates it rather than copying it. The structure we use (abridged):

@Image1 is the first frame (0s) and @Image2 the last frame (10s) of one
continuous shot: two moments of the same take, never blend them.
@Video1 is a split-frame turntable character sheet: use it only to keep
face, hair and clothes identical from every angle, including from behind.

Locked-off eye-level side camera, 10 seconds, no cuts. Shallow depth of
field: car and man sharp, house and trees in soft creamy bokeh.
The car is parked and completely stationary from first frame to last.

0-2s   he sits at the wheel, hand on the wheel, gazing ahead
2-3s   he leans toward the door, which starts to open
3-5s   door fully open, he steps out onto the dirt
5-6s   he stands beside the open door, back to camera
6-7s   he shuts the door
7-10s  he steps forward and leans on the car, as in @Image2, and holds

Face and hair identical to @Video1. No glasses, no extra people.

The time-coded beats were read off the original shot, not guessed. The prompt must keep the vehicle still explicitly, and it stays under the length limit of the tool (2,500 characters).

5Results

The original on the left of the line, a refocused version on the right. Move over the picture to compare, and switch between the two video models: Seedance 2.5 and Kling 4.

Original
0:00 / 0:00
Original
Original
Seedance 2.5
Seedance 2.5
Kling 4
Kling 4

The same shot at 5 s: original, Seedance 2.5 and Kling 4. The video comparison is in the web version of this note.

VersionLengthObservation
Original10.1 sFocus on the house, the car and the man soft. Action: seated, door opens, steps out, stands, shuts the door, leans on the car.
Seedance 2.510.0 sSame sequence and same final pose, with the refocus applied from start to end. Details of the movement differ.
Kling 410.0 sSame sequence and same final pose. The door and the body movement differ from the original at the same moments.

Both versions are very close to the original action, but not identical to it. We compared them by eye on the video; we have not measured the timing difference.

6What did not work

SymptomCauseResponse
Intermediate keyframes made the video worseThe more keyframes that look alike, the more the model morphs between themKeep only the first and the last frame
Refocused frames are not identical to each otherEach Flux generation is uniqueNot an issue inside one shot. Unsolved between neighbouring shots
Face changes during the shotThe face is out of focus in the frames, no clean referenceFull-body image, character sheet, 360° turnaround video in the prompt
Action not exactly the originalThe shot is recreated, not editedTime-coded prompt read from the original. Close, not identical

7Limits and what we would change

References

  1. Black Forest Labs, FLUX. Image model used to refocus the first and the last frame.
  2. Flora, Character 360 kit. Technique used to make the character sheet and the 360° turnaround video.
  3. Flora. The canvas where the references were built.
  4. ByteDance Seed, Seedance 2.0. Video model, first and last image with a reference video.
  5. Kling AI, Kling 4. Video model, first and last image with reference images.

Credits

Written by Arno Faure, Director of Creative Technology and Innovation.