Stadium Calm Down: the only calm one in a roaring stadium, starring you

How to make a Stadium Calm Down video yourself

You do not need this page to make one. This is how the video is put together, with the prompts for every shot, so you can build it with any picture model and image-to-video model you like.

1. Step 1

   Draw the opening from your photo, already in the suit

   Use a picture model that takes a reference photo and draw the opening still close up, from the chest up: you side-on at the very front of packed stands, looking back at the camera over one shoulder. Describe the suit and the stands in full, and say there is no sky, pitch, floodlight or advertising board in the picture and no crest, badge, word or number anywhere.

2. Step 2

   Draw the other three from the first one

   For the walk, the presses at the fans and the look into the lens, give the picture model your photo and the opening still as a second reference, and ask it to keep the suit and the stands from that still but draw the framing you describe: from the thighs up, from the waist up, then from the chest up, with both palms held flat and facing down.

3. Step 3

   Animate each still and cut them together

   Give each still to an image-to-video model and spell out the move: the turn first, then the presses, each the width of a hand, about one a second, with the fans behind still jumping and shouting. Trim the turn to 3.5 seconds, the walk to 4.5, the presses at the fans to 3.5 and the look into the lens to 4.5, lay them end to end at 9:16 with hard cuts, and add a slow, soft sound on TikTok.

The shots, second by second

The prompts, shot by shot

Each shot is made in two steps. Give its still prompt to a picture model with your photo as the reference, then give the picture it draws to an image-to-video model with the motion prompt. Add these endings to every prompt: they set the look and keep the faces the same from shot to shot.

1. 0–3.5sThe turn

   Still prompt

   Copy

   A close shot from the chest up: the head and shoulders of the person in the reference photo, wearing a plain dark charcoal suit, jacket and trousers, over a plain white shirt with an open collar and no tie, fill the frame. They stand side-on at the very front of the stands at a packed football stadium during a night match. Right behind them, a dense wall of fans fills the whole background edge to edge, on their feet, roaring, mouths wide open, arms raised and waving, in plain T-shirts of solid white or solid blue with nothing on the chest or anywhere else, lit by bright white stadium light. No sky, pitch, floodlights or advertising boards are in the picture. Nobody holds a flag, scarf, sign or banner, and nothing has a crest, badge, logo, word, number or letter on it. Their body is turned toward the fans, and they look back over one shoulder at the camera, their face turned to it and in full view, with a cool, unbothered, deadpan look, mouth closed. They do not smile.

   Motion prompt

   Copy

   Immediately they turn the rest of the way round to face the camera, slow and unhurried, hands loose at their sides, with a cool, unbothered look, mouth closed. They do not smile. The fans right behind them keep jumping, shouting and waving their arms the whole time, in plain T-shirts with nothing on them. The camera holds steady, a little handheld. It stays the same night match under the same bright stadium light throughout: the light does not change.

2. 3.5–8sThe walk

   Still prompt

   Copy

   A medium shot from the thighs up, from close in front of them: their head, body and both hands fill most of the frame, and their knees and feet are out of the picture. The person in the reference photo, wearing the same suit and shirt as in the other frame, faces the camera at the very front of the stands, as in the other frame: a dense wall of fans fills the whole background edge to edge, from the bottom of the picture to the top, in plain T-shirts with nothing on the chest. No sky, pitch, floodlights or advertising boards are in the picture. Nobody holds a flag, scarf, sign or banner, and nothing has a crest, badge, logo, word, number or letter on it. Both hands are held out low in front of their hips, palms flat and facing the ground, fingers together. Face in full view, with a cool, unbothered, deadpan look, mouth closed. They do not smile.

   Motion prompt

   Copy

   They walk slowly toward the camera, unhurried, and press both palms down in front of their hips again and again, the width of a hand down and back up, about one press a second: the calm-down gesture. The camera backs away ahead of them at the same pace, a little handheld, keeping them from the thighs up. Their face stays cool and unbothered, mouth closed. They do not smile. The fans right behind them keep jumping, shouting and waving their arms the whole time, in plain T-shirts with nothing on them. It stays the same night match under the same bright stadium light throughout: the light does not change.

3. 8–11.5sAt the fans

   Still prompt

   Copy

   A close shot from the waist up: their head, chest and both hands fill the frame, and everything below their belt is out of the picture. The person in the reference photo, wearing the same suit and shirt as in the other frame, stands at the very front of the stands, as in the other frame: a dense wall of fans fills the whole background edge to edge, from the bottom of the picture to the top, in plain T-shirts with nothing on the chest. No sky, pitch, floodlights or advertising boards are in the picture. Nobody holds a flag, scarf, sign or banner, and nothing has a crest, badge, logo, word, number or letter on it. A steward in a plain yellow high-visibility vest with nothing on it stands behind them, back to the camera. Both hands are out in front of their waist, palms flat and down; they glance sideways at the fans, one eyebrow raised, face three-quarters to the camera, in full view, with a cool, unbothered, deadpan look, mouth closed. They do not smile.

   Motion prompt

   Copy

   Immediately they press both palms down toward the roaring fans beside them, slowly, the width of a hand down and back up, three times, as if telling the whole stadium to calm down, glancing at the fans with one eyebrow raised, then look back at the camera. Their face stays cool and unbothered, mouth closed. They do not smile. The fans right behind them keep jumping, shouting and waving their arms the whole time, in plain T-shirts with nothing on them. The camera holds steady, a little handheld. It stays the same night match under the same bright stadium light throughout: the light does not change.

4. 11.5–16sInto the lens

   Still prompt

   Copy

   A close shot from the chest up: their head, shoulders and both hands fill the frame. The person in the reference photo, wearing the same suit and shirt as in the other frame, faces the camera square on at the very front of the stands, as in the other frame: a dense wall of fans fills the whole background edge to edge, from the bottom of the picture to the top, in plain T-shirts with nothing on the chest. No sky, pitch, floodlights or advertising boards are in the picture. Nobody holds a flag, scarf, sign or banner, and nothing has a crest, badge, logo, word, number or letter on it. Both hands are raised side by side in front of their chest at the same height, palms flat and facing down, fingers together. Their face is in full view, looking straight into the lens, with a cool, unbothered, deadpan look, mouth closed. They do not smile.

   Motion prompt

   Copy

   They press both palms down slowly, the width of a hand, three times, looking straight into the lens with a cool, faintly knowing look, then give one slow nod. Mouth closed. They do not smile. The fans right behind them keep jumping, shouting and waving their arms the whole time, in plain T-shirts with nothing on them. The camera holds steady, a little handheld. It stays the same night match under the same bright stadium light throughout: the light does not change.

Where it goes wrong

• Start with the face, not the back of the head

  The original opens with his back half to the camera. Ask a video model to turn someone round from behind and it has to make the face up, and it often comes round as someone else. Start side-on with the face already looking back over the shoulder, and put the rest of the turn at the very start of the motion.

• Give the press a size and a pace

  Told only "the calm-down gesture", a video model barely moves the hands. Say the palms are flat and facing down, that each press goes the width of a hand down and back up, about one a second, and how many there are, and draw the still with the hands already out in front.

• Keep crests and names out of the crowd

  Say "football stadium" and a picture model hangs flags and banners with words on them, puts crests and sportswear marks on the shirts and names and numbers on the backs. Say in every still that nobody holds a flag, scarf, sign or banner and that nothing has a crest, badge, logo, word or number on it, and name the chest: plain T-shirts with nothing on the chest. Told only "plain tops", one of our test stills still put a maker's tick on a fan's shirt. Leave out the player walking past.

• Keep the camera right in front of the stands

  Say the person is a few metres from the camera, or that it films from pitch level, and a picture model pulls back to the whole stadium: the pitch, the floodlights and lettered advertising boards, which happened to one of our test stills. Put the camera close in front of them and say the fans fill the picture down to its bottom edge.

• Keep the straight face

  Left alone, a model makes people smile, and one of the AI remakes ends on a big grin, which kills the joke. Say "They do not smile" in every still and every motion, with a cool, unbothered look and the mouth closed.

• Only people who said yes

  Use your own photo or one of someone who agreed. Do not use celebrities, the man from the original clip, or anyone who has not agreed to be in the video: a realistic video of a real person in a stadium they were never in, made without their consent, can hurt them, and platforms take it down.

• Label it as AI-generated

  Switch on TikTok's AI-generated content label before you post. A clip that looks like live sports coverage is exactly the kind people take at face value, and viewers trust a post that says so.