The exact prompts, models, and audio pairing to recreate the trend that's taking over your FYP.
If you've been on TikTok or Reels in the last few weeks, you've seen her. A stunning fan caught on the stadium fan-cam mid-bite, mid-smile, mid-game. The whole crowd watching. The internet replaying it on loop.
Here's the part nobody's telling you: she isn't real. She's AI. And she's everywhere because she works. That perfect intersection of candid realism, viral broadcast aesthetic, and a soundtrack that punches you in the chest.
This guide walks you through the exact three-prompt stack to recreate her, start to finish, using ChatGPT for the stills and Higgsfield for the video. No filler. Just the prompts, the settings, and the sound.
Three prompts. Two tools. Fifteen seconds of footage that looks like it was ripped straight off a FOX Sports broadcast.
Open ChatGPT. Upload a clear, well-lit photo of yourself. Face forward, good lighting, minimal filtering. This becomes your identity anchor for everything that follows. Paste in Prompt 01: The Stadium Sit. Generate. This gives you the wide crowd-cam shot of you seated in the stands.
Stay in the same ChatGPT chat so it keeps your identity locked. Paste in Prompt 02: The Hot Dog Bite. Generate. You now have your second still: you, eating the hot dog, fan-cam framing.
Go to Higgsfield → Select Video → choose Seedance 2.0 as your model. Upload both images you generated in Steps 1 and 2.
Paste in Prompt 03: The Video Brief. Reference both images directly inside the prompt. This is non-negotiable — without the @references the video won't hold identity continuity. Then click Generate.
Drop these into the tools as written. Replace the team placeholder. Don't paraphrase — the specific language is doing structural work.
Analyze the attached reference image and preserve the exact identity, facial structure,
hairstyle, skin tone, ethnicity, proportions, and realism of the person in the reference
image.
Create a realistic live sports broadcast crowd shot of the person in the reference image
sitting naturally at a packed [INSERT SPORTS TEAM HERE] game inside a professional stadium.
The person should look casually attractive and naturally confident, like a real spectator
accidentally shown on a stadium fan camera during the live broadcast.
The person is seated comfortably watching the game, not posing for the camera. Their posture
should feel relaxed and natural while looking toward the field or gameplay action.
Subtle realistic styling enhancements:
− polished realistic hair with natural texture
− healthy skin texture with pores and imperfections
− subtle makeup and lip gloss
− naturally flattering sports-model appearance
− realistic outfit styling inspired by the team
− flattering but believable stadium lighting
DO NOT:
− change the person's identity
− create AI beauty-face aesthetics
− reshape the face
− enlarge eyes unnaturally
− create unrealistic curves or proportions
− over smooth skin
− create influencer-style glam posing
− use cinematic Hollywood grading
− overly sharpen the image
Environment: authentic packed stadium crowd, realistic fans nearby, jerseys, drinks, food
trays, phones, realistic stadium atmosphere, scoreboard lighting glow, slight motion blur,
broadcast compression artifacts, candid telephoto sports camera framing, live TV screenshot
quality, documentary realism, imperfect framing like a real sports broadcast.
Camera style: telephoto sports lens look, shallow depth of field, realistic exposure and
shadows, horizontal 16:9 composition, authentic live broadcast camera feel.
Final image should feel like: "A naturally attractive real person accidentally captured on a
live sports broadcast who later went viral online." FOX SPORTS broadcast aesthetic.Analyze the attached reference image and preserve the exact identity, facial structure,
hairstyle, skin tone, ethnicity, proportions, and realism of the person in the reference
image.
Create an ultra realistic live sports broadcast crowd shot at a packed [INSERT SPORTS TEAM
HERE] game inside a large professional stadium during a night game.
The person in the reference image is seated naturally among fans wearing a fitted stylish
[INSERT SPORTS TEAM HERE] jersey/top inspired by the team colors and branding. The styling
should feel naturally attractive, polished, and photogenic while still looking completely
realistic and human.
The person is casually eating a stadium hot dog while focused on the game. The expression
should feel candid and authentic, like a real fan accidentally captured by a live TV fan
camera during the broadcast.
Subtle realistic enhancements only:
− healthy glowing skin texture
− realistic hair styling with natural flyaways
− subtle makeup and lip gloss
− refined but believable facial detail
− flattering stadium lighting
− naturally stylish sports-fan aesthetic
− realistic wardrobe fit and silhouette
DO NOT: change identity, reshape facial features, create giant eyes, over smooth skin, create
unrealistic body proportions, make the image look AI generated, create influencer glam posing,
add cinematic Hollywood color grading, overly sharpen the image.
Environment: packed realistic stadium crowd, fans cheering naturally, beer cups, hot dogs,
trays, phones, foam fingers, authentic sports atmosphere, realistic scoreboard lighting glow,
slight motion blur, broadcast compression noise, candid telephoto sports camera framing, live
TV screenshot feeling, documentary realism, slightly imperfect framing like a real fan cam
moment.
Camera style: telephoto sports lens look, shallow depth of field, realistic shadows and
exposure, horizontal 16:9 framing, authentic live broadcast capture quality.
Final image should feel like: "A naturally attractive real person accidentally captured on a
live sports broadcast who later went viral online." FOX SPORTS broadcast aesthetic.Use @image_1 as the main action and pose reference. Use @image_2 as the exact identity and
beauty reference. Keep the same woman consistent throughout the entire video: same face,
hair, eyes, jersey, necklace, skin tone, body proportions, and overall appearance.
Create a realistic 15-second viral MLB broadcast-style video in 9:16 vertical format,
optimized for TikTok and phone viewing.
The scene is a packed baseball game in the grandstands, styled like a real live TV broadcast
crowd camera moment. The woman is sitting naturally among fans, wearing the same jersey and
necklace from the reference images. She is holding a classic stadium hot dog.
The video begins with her watching the baseball game intently. She looks focused, calm, and
serious, fully absorbed in the action on the field. She is not posing and has not noticed
the camera yet.
At the very start of the clip, she takes a large, real bite from the hot dog. Make this first
bite clearly visible and natural. After the bite, the hot dog should remain half-eaten with
a visible bite mark for the rest of the scene, so the continuity stays realistic.
The live broadcast camera slowly pans across the crowd and lands on her. It gently zooms in
like a real sports broadcast operator has spotted her in the audience. She is caught just
after taking that big bite, chewing naturally, completely unaware and candid.
Then she notices herself on the jumbotron. She pauses mid-bite, slightly embarrassed, then
slowly smiles. Her expression changes from focused to surprised to playful confidence. She
gives a soft, charming laugh and holds eye contact with the camera for a brief moment, like
she knows the whole stadium is watching.
Keep the moment attractive, confident, and emotionally engaging, but still realistic and
candid. She should look beautiful and magnetic, but not fake, not overly posed, and not like
an AI influencer. The appeal should come from her natural reaction, her confident smile, the
live crowd energy, and the feeling of being accidentally caught on camera.
People around her begin to notice, smile, point, and react naturally. The stadium atmosphere
builds around her. The shot should feel like a real viral broadcast clip that people would
replay because her reaction is funny, charming, and memorable.
Visual style: realistic MLB live broadcast, FOX Sports-style crowd camera, documentary
realism, telephoto sports lens, natural stadium lighting, subtle motion blur, slight broadcast
compression, authentic crowd movement, real game atmosphere, candid viral sports moment.
Camera: vertical 9:16 framing, slow pan into a medium close-up, gentle broadcast-style zoom,
keep her centered, face sharp, hot dog visible with the bite mark, crowd slightly compressed
in the background.
Audio: include mild natural stadium background noise: distant crowd chatter, light cheering,
ambient ballpark atmosphere, subtle game-day noise. Keep the audio realistic and not
overpowering.
Important: do not change her identity. Do not change her face, hair, jersey, necklace, or
body proportions. Do not make her look like a different person. Do not make it look like a
deepfake. Keep her realistic, natural, and consistent with the reference images.
Do not include a scoreboard overlay. Keep only a small broadcast-style bug/branding element
on screen, with no large graphics or scoreboard taking attention away from the woman.
Continuity lock: maintain prop continuity. After the first big bite, the hot dog must remain
visibly half-eaten with the same bite mark throughout the rest of the clip.The version that goes viral is the one with your face, your team, your caption, your moment.
Tag @RealMissAI when you post it.
← Back to vault