Use the attached image as the primary character reference

English Prompt

Use the attached image as the primary character reference.

Create an ultra-realistic live-action video that looks like an authentic baseball TV broadcast accidentally cutting to a fan in the stadium audience.

Scene: a crowded professional baseball stadium at night. The atmosphere feels real, noisy, and energetic, with stadium lights, blurred spectators in the background, cheering goods, team colors, and a slightly compressed broadcast-camera look.

Main subject: keep the same woman from the reference image as closely as possible, including her green bob hair, crystal horn headband, soft pale skin, pinkish eyes, natural makeup, green baseball fan jersey, and cheering goods. She is sitting in the audience, holding a drink and inflatable cheering sticks.

Action:
At first, the stadium spotlight lands on her, but she does not notice it. She keeps looking toward the field with a calm, slightly serious expression, as if she is focused on the game. The TV camera slowly zooms in from a medium crowd shot to a closer broadcast shot. People around her start noticing that she is on the big screen and on the TV camera. A nearby fan gently points toward the camera and tells her where to look. She looks confused for a moment, then realizes the camera is filming her. She becomes shy and slightly embarrassed, gives a small nervous smile, then softly waves her hand toward the camera. Her smile should feel natural, modest, and cute, not exaggerated.

Camera style:
Realistic sports broadcast footage, telephoto lens, shallow depth of field, handheld broadcast camera movement, slight camera shake, natural zoom, realistic stadium lighting, authentic crowd motion, cinematic but believable. The subject should be naturally lit by the stadium spotlight while the background remains softly blurred.

Visual details:
Add subtle TV broadcast elements such as a scoreboard overlay, inning information, and game-style lower-third graphics, but keep them visually realistic and not distracting. The footage should look like a real Japanese baseball broadcast or live sports TV capture, not a staged photoshoot, not a music video, not anime, not CGI.

Mood:
Charming, candid, slightly funny, wholesome, and very realistic. The moment should feel like a genuine fan-cam moment during a live baseball game.

Negative prompt:
anime style, illustration, CGI, 3D render, overly perfect beauty shot, fashion photoshoot, smartphone selfie angle, unrealistic camera movement, distorted face, extra fingers, broken hands, unreadable messy text, fake-looking crowd, exaggerated acting, dramatic slow motion, low-quality face, plastic skin, unnatural smile.