Prompt

Use the uploaded image as the ONLY identity reference. Strictly preserve the subject’s facial features, structure, proportions, texture, and overall identity, while naturally adapting lighting and perspective to the stadium environment. The subject must remain fully consistent with its original species at all times, with no anthropomorphic transformation or human-like reinterpretation. The appearance must not look pasted or overfitted. A subject within a dense stadium crowd is captured by a live broadcast camera. The subject is positioned naturally within the environment, integrated among surrounding spectators. INITIAL ACTION (EATING): At the beginning of the shot, the subject is actively eating, taking a large, slightly exaggerated bite of food. The food is oversized but realistic. The action is mid-motion and unposed, like casually eating while watching the match. DISCOVERY LOGIC (SEE SCREEN FIRST — CRITICAL): The subject does NOT notice the camera directly. Instead, the subject briefly turns their head to the side and looks toward .On the screen, they recognize themselves being shown live. REACTION SEQUENCE (VERY FAST & NATURAL): The reaction happens immediately in a tight sequence: A quick, subtle look toward the big screen A tiny moment of recognition (less than half a second) A sudden, light “caught off guard” reaction (very slight startle, not exaggerated) Instantly transitioning into a shy, slightly embarrassed smile FOOD ACTION (SIMULTANEOUS, SLIGHTLY FLUSTERED): At the SAME TIME as the shy smile appears, the subject quickly and instinctively lowers the food: the food is pulled away from the mouth a bit too fast there is a small hesitation or micro-adjustment the subject places or lowers it toward their lap or side in a slightly messy, hurried way The movement must feel unplanned and slightly clumsy, like reacting while being watched. FACIAL + BODY LANGUAGE: The smile is soft, shy, and a bit embarrassed, appearing naturally after the realization. The subject avoids direct eye contact with the camera, looking slightly away or downward after noticing. Shoulders draw in slightly, posture tightens briefly, then settles. AUDIO (STRICT HARD RULE): Only real stadium ambient sound: continuous crowd noise, distant cheering, layered audience murmur. ABSOLUTELY NO chewing sounds, no mouth sounds, no eating sounds, no laughter, no speech, no dialogue, no music. MOTION STYLE: All motion must be subtle, continuous, and realistic, like a candid live TV broadcast capture. Include slight broadcast micro-shake, natural crowd movement, and imperfect timing. The final result must feel like an authentic, slightly awkward, candid sports broadcast moment — not staged, not cinematic, and not stylized. Exclude direct eye contact with the camera, exaggerated acting, staged comedy, unrealistic physics, cartoon/anime styles, or any non-broadcast composition.