ask your agent this
Need you to do 2 things with this image and I'll wait for the final results.
1 - Using GPT IMG 2 : Use the main subject in the reference photo as a stunningly realistic screenshot from a live World Cup (soccer) match. The camera pans to the crowd, focusing on a person in the reference image, seated in the front row VIP section with a natural smile, seemingly unaware that they are being broadcast live. Please preserve the person's facial structure, identity, and portrait as accurately as possible.
The match is a semi-final, US vs. Brazil, with Brazil currently leading on aggregate. The image realistically recreates the stadium atmosphere: packed stadiums, bright lights, waving scarves, and blurred figures of fans in the background. The composition of the figures is like a real crowd reaction captured by a live broadcast camera.
It includes all the elements of a live football broadcast overlay: live score, match timer, sports channel watermark/logo, the bottom third of the match, and a true 16:9 television aspect ratio. The image should feel like a real paused television broadcast—realistic sports color gradation, slight motion blur, subtle compression artifacts, interlaced grain, telephoto lens compression, cinematic broadcast realism, and natural stadium camera composition.
after you have the Image proceed with step 2
2: Use Seedance 2 to animate the image following the prompt (--15sec) :
It’s a realistic live World Cup soccer match broadcast during a USA vs BRAZIL Semi-Final match at a modern stadium. The camera cuts to the subject sitting in premium front-row seats near the pitch, like a genuine live TV crowd reaction shot featuring a notable guest in attendance. The subject smiles naturally and casually without over-performing for the camera. Subject mainly watches the match, occasionally glancing toward the jumbotron or briefly toward the camera before returning her attention to the field. Preserve the subject’s identity and facial structure perfectly.
One continuous take with no cuts and no angle changes.
From 0–4 seconds, the camera settles on her while Subject watches the match attentively with a relaxed smile.
From 4–7 seconds, Subject leans forward slightly with excitement and claps naturally once or twice like a real football fan reacting to the intensity of the match.
From 7–11 seconds, Subject noticed was being photographed. and gives a relaxed thumbs-up and a brief casual wave toward the broadcast camera. Nearby supporters cheer louder when they recognize her. Subject glances up at the jumbotron, smiles, then looks back toward the field.
From 11–15 seconds, she laughs naturally while silently saying something to the friend beside her, then joins the crowd with light applause and subtle participation in the stadium chant while remaining focused on the match.
Broadcast styling should feel exactly like a real World Cup television broadcast. Include a realistic soccer scorebug at the bottom of the screen that stays completely unchanged throughout the entire 15 seconds. Do not animate or update it. Use a genuine USA vs Brasil broadcast layout. Above the scorebug, display a clean lower-third graphic reading: “Made by Pika Agent at
pika.me,” styled like a real football broadcast guest identifier.
Audio should feature natural live football-broadcast commentary from two male commentators casually discussing Subject appearance at the match in a warm, authentic tone. Background audio includes massive stadium crowd ambience, football chants and cheers, distant referee whistles.