How to Make DLS Style AI Videos Step by Step

The step-by-step guide to creating DLS-style AI videos



This is the easiest workflow to use in case you would like to create Dream League Soccer-style AI videos such as broken goalposts, injury scenes, funny medical rescues, celebrations, or crowd moments.

It is not necessary to do everything in a single tool. The easiest way is:

Image prompt > create image in MagicAI > download image > continue it in ChatGPT and ask it to animate Image prompt > edit in CapCut or KineMaster

That is the full process.

Step 1: Decide the scene first

Since you are going to write something, you have to know your scene.

For example:
  • broken goalpost scene
  • helicopter rescue scene
  • medical staff funny fat scene.
  • celebration with fireworks
  • iftaar tables on the pitch
  • referee between them arguing with players.

Do not get down into animation at once. First lock the idea.

An excellent scene concept must respond to three questions:
  • What is happening?
  • Where is it happening?
  • Who is in the frame?

Example:

A player of the yellow kit is wounded in the middle, the teammates and the rivals are surrounding him, and a medical worker is slowly running to the field.

That is a complete scene idea.

dls ai video

Step 2: Take or select a reference screenshot.


Now put a clean gameplay shot of your DLS-style game.

Attempt to select a screenshot with:
  • A stable camera angle
  • Playable UI in case you desire the game feel.
  • Room in the frame to develop the new action.
  • Distinct players and markings of pitch.

This is the screenshot that you will use as your image of reference.

That is important since what you should remember about your prompts are:

  • The same camera angle
  • The same stadium layout
  • The same field lines
  • The same UI

This is what causes the end result to appear like actual gameplay rather than artificial AI.

Step 3: Write the image prompt


Write a prompt to the still image first.

To include in your image prompt you should always have:

  • Reference to uploaded image.
  • Do not change camera angle
  • Keep gameplay UI visible
  • Describe the scene clearly
  • Explain positions of people.
  • Describe mood or reactions

The simplest structure would appear as follows:

Image prompt structure

Take the uploaded image as a reference image. DO NOT alter the camera, stadium map, pitch and the gameplay user interface. Retain the same Dream League Soccer gameplay point of view.

Scene: [describe the scene in which you are]

Notice: The camera angle during the gameplay and all the UI elements should be kept exactly the same as in the reference picture.

Example image prompt

Take the uploaded image as a reference image. DO NOT alter the camera, stadium map, pitch and the gameplay user interface. Retain the same Dream League Soccer gameplay view.

Setting: On the field, one of the players is lying on the ground because he is injured. There are a number of his teammates and some opponent players standing around him in a state of concern. A member of the medical staff in red rescue uniform is walking towards the injured player with a first-aid box on the sideline. Every player around the injured player is facing towards the medical staff approaching.

Notice: The camera angle during the gameplay and all the UI elements should be kept exactly the same as in the reference picture.

dls ai video

Step 4: this is also pasted in MagicAI

MagicAI is already open so now select the image generating option.

Then:
  • Post your screenshot of the gameplay
  • Paste your image prompt
  • Generate the image
  • Make variations if needed

Do not rest content too soon with the first result. The second or the third attempt is usually better.

When creating: continue checking:

  • Was the angle of the camera the same?
  • Did the UI remain visible?
  • Do the players play where they belong?
  • Is the main action clear?

In case it is not right, re-create and repeat.

Step 5: The best generated image is downloaded.


When the image appears to be fine, download it.

This is significant since the created image is the foundation of animation.

In case your new scene is already created in the AI image, you should not animate the original screen shot.

The best workflow is:

Original screen shot > scene image generated by AI > animated video.

Step 6: Re-upload that image generated into ChatGPT.


It is time to go back to ChatGPT and post the final generated image.

The next thing is an animation prompt request.

This is where the majority of the people go wrong. They attempt to bring to life without a motion plan.

Rather, prompt ChatGPT to write an animation prompt with:

  • Fixed camera
  • Fixed UI
  • Exact movement timeline
  • What comes and comes and comes and comes.

A powerful request will appear as follows:

Write a copy-paste 6 seconds animation cue on this image. He maintain the same camera angle and UI. The medical staff moves out of the sideline, goes to the injured player, kneels, and begins treatment.

That provides ChatGPT with sufficient structure to create a prompt to use.

Step 7: Get animation prompt of ChatGPT.


An effective prompt of animation should entail:
  • Reference to uploaded image.
  • Keep same camera angle
  • Keep same UI
  • Explain the scene
  • Divide the action into the seconds.

Animation prompt structure

Take the uploaded image as the reference image. Retain the identical gameplay camera perspective, stadium map, pitch markings as well as mobile game interface.

Scene: [describe the scene]

Animation sequence (Number of seconds animated):

0-2 seconds: [first movement]
2-4 seconds: [second movement]
4-6 seconds: [final movement]

Noteworthy: UIC should not be changed, and the camera angle should remain constant.

Example animation prompt

The uploaded image is used as the reference image. Retain the identical Dream League Soccer gameplay camera view, stadium map, pitch markings and mobile game interface.

Setting: There is one player on the grass in the middle of the pitch, lying wounded. A few opponents and a number of teammates are around him in concern. One of the medical staff members in a rescue uniform is coming to the side with a first-aid box.

Animation frame rate (6 seconds duration):

0-2 seconds: The medical personnel slowly walks along the sideline to the injured player and other players in the same team stare at her.

2-4 seconds: She comes to the injured player, crouches next to him and puts the first-aid box on the grass.

4-6 seconds: The players will remain on the spot with the body movements of concern as the medical personnel opens the box and begins treatment.

Notable: Do not change the camera angle or the interface of the gameplay. Do not move or zoom the camera.

Step 8:Enter the animation prompt into MagicAI.


We are now going to go back to MagicAI and select the animation or image to video option.

Then:

  • Post the still image that has been obtained.
  • Paste the animation prompt
  • Choose the duration
  • Generate the video

The following are some of the most common small errors that occur at this stage:

  • Camera movement
  • Disappearing players
  • Cchanging UI
  • Wrong movement speed

In case MagicAI is altered too much, add more powerful instructions such as:

Keep the same camera angle.
Do not move the camera.
Keep all UI elements visible.
And animate only those things and people that have been described.

That helps a lot.

Step 9: There is a downloading of the animated clip.


After making sure the motion is good, download the clip.

Although the outcome may not be ideal, save all the versions that are usable. There are cases whereby one generation possesses superior player motion and the other has superior background stability.

One can always choose the best during editing.

dls ai video

Step 10: Finalize the video in CapCut, KineMaster or any other editor.


At this point, insert the animated clip in one of your editing programs.

You can use:
  • CapCut
  • KineMaster
  • VN
  • Alight Motion
  • Premiere Pro
It is possible to enhance the clip in the editor with:
  • Sound effects
  • Crowd noise
  • Dramatic music
  • Text overlays
  • Slow zoom in post
  • Subtitles
  • Transitions between scenes
As an example, when you made a helicopter rescue clip, you can add:

  • Helicopter rotor sound
  • Crowd gasp sound
  • Dramatic background music
  • Such text as Emergency on the pitch.
In case it is a party video, include:
  • Fireworks sound
  • Cheering crowd
  • Goal horn
  • Confetti overlay

It is at this point that the AI clip is transformed into an actual short video.


Step 11: cut up several clips into a complete story.

One clip is good. Several interrelated clips are far better.

As an example, you can tell your whole story:

  • Player gets injured
  • Players gather around
  • Medical staff enters
  • Rescue scene happens
  • Tribute scene later
Or:
  • Goal is scored
  • Celebration of fireworks begins.
  • Crowd cheers
  • Players dance
  • Final freeze frame

That is the way you make viral football story video out of random ideas.

Step 12: Work with the same workflow on a regular basis.

The simplest formula that can be repeated is:

  1. Full workflow
  2. Choose the scene
  3. Take gameplay screenshot
  4. Write image prompt ChatGPT.
  5. Paste into MagicAI
  6. Generate image
  7. Download image
  8. Upload image back to ChatGPT
  9. Ask for animation prompt
  10. Prince paste animation Activate MagicAI.
  11. Generate video
  12. Download video
  13. Edit in CapCut or KineMaster

That is the whole pipeline.

Best prompt tips

In order to achieve improved outcomes, you should always add the following lines in your prompts:

For image prompts

Take the uploaded image as the reference image.
Please, do not alter the camera angle.
Maintain same stadium plan and pitch decorations.
Keep the gameplay UI visible.

For animation prompts

Same camera angle of the gameplay.
Do not zoom or move the camera.
Keep all UI elements visible.
Just bring to life the mentioned objects and characters.

Most of the problems are solved by these four lines.

Common mistakes to avoid


Most novices spoil the outcome due to the fact that they:

  • Write very short prompts
  • Forget to mention the camera angle
  • Forget to mention UI
  • Animate the source image rather than the generated image.
  • Do too much in too little time.
  • Do not use a timeline

Always remember:

Image first: Animation second. Editing third.

That order matters.

Final advice

It is not only about using AI.

The trick is that AI should be used in the correct sequence:
  1. First for the still scene
  2. Then for movement
  3. Then for editing and sound
This is how such DLS-style videos are produced in a clean manner.

When you jump over steps, the outcome is disheveled.
When you do it, you can make your videos appear much more professional.

Post a Comment

Previous Post Next Post