A candid photo that looks like REAL handheld smartphone footage taken from inside a crowd at an outdoor event, slightly unsteady, shot from the audience over people's heads. Scene: an outdoor performance stage recreating a Japanese ninja-village gate, set against forested green mountains under natural daylight — a tiled-roof wooden building, a grey stone wall bearing a black banner with a white three-comma "tomoe" crest, tall purple vertical banners, a tall wooden pole, and a large white cloth with a single black Japanese kanji on the wall. A large audience in the foreground holds up smartphones to film the show. Two cosplay performers stand center stage in wide low martial-arts fighting stances, facing each other across the stage. The LEFT performer wears a full white martial-arts ninja outfit with white boots and a white belt, and IS the person in the FIRST reference image — keep their exact face, hair, body and GENDER; if the reference is a woman the left performer is a woman, if a man then a man — do not change them into a different person or gender. The RIGHT performer wears an all-black ninja outfit with dark boots, and IS the person in the SECOND reference image — keep their exact face, hair, body and gender. Natural daylight, realistic colors, slight digital noise and phone-camera grain, shallow depth of field with the stage in focus and crowd silhouettes softer in the foreground. Photo-real, documentary, candid — NOT a clean studio render. No text overlays, no captions, no watermark.
Free to start · Generate videos and images with AI in seconds