Create a 13-second vertical 9:16 authentic iPhone-front-camera UGC summer try-on video. @Image2 is the sole presenter identity reference: preserve her recognizable face, warm complexion, athletic-slim proportions, blonde hair color and sleek high ponytail consistently across every shot. Do not retain the beige trousers, white sneakers, original wardrobe or studio background from @Image2.
@Image1 is the absolute outfit reference. Reproduce the loose ivory-white asymmetric short-sleeve top with its wide neckline, visible sage-green camisole straps, two watercolor-style green bow graphics, small outlined green stars and tiny heart details; the soft sage-green pull-on shorts with an elastic gathered waistband, cream drawstring tied in a bow, slanted side pockets and double ruffled hems; and the cream flat flip-flop sandals with pale floral fabric straps tied into soft knots above the toes. Do not convert the shorts into a skirt, remove the ruffles or pockets, change the graphic placement, turn the sandals into sneakers or heels, or add unsupported jewelry and accessories.
Use one continuous bright, cozy bedroom or apartment dressing area with pale neutral walls, soft natural window daylight, a full-length mirror and a few believable lived-in details. Keep the environment tidy and casual rather than commercially staged. Render realistic skin and fabric texture, stable natural exposure, subtle autofocus breathing and mild handheld smartphone movement. No beauty filter or stylized color grading.
0.00–3.40 seconds: Medium direct-to-camera framing. The woman wears a simple fitted white tank and soft neutral lounge shorts while holding the @Image1 ivory bow-print top open with both hands. Keep the asymmetric neckline, sage straps, both green bows and the small star details readable. She smiles with pleasantly surprised “this looks effortless” energy and says with accurate lip sync: “Okay, this might be the easiest summer outfit ever.” Let the cotton top move with natural lightweight fabric behavior and keep both hands anatomically correct.
At 3.40 seconds, use a clean hard cut—no morphing or continuous dressing—to the complete outfit already worn.
3.40–8.30 seconds: Full-body frontal view in the same room. She now wears every @Image1 component with source-faithful construction. The ivory top falls loosely over the sage shorts, the green camisole straps remain visible, and the waistband, cream drawstring, side pockets and double ruffled hems are clear. She adjusts the asymmetric neckline once, reties the drawstring into a neat bow and lightly brushes one ruffled hem. She steps back and says with genuine delight: “The sage shorts with the little bows? I’m obsessed.”
At 8.30 seconds, use a second clean hard cut to a slightly wider head-to-toe view in the same location.
8.30–13.00 seconds: Keep her head, both ruffled hems and both sandals fully inside frame. She places one foot forward so the cream soles and floral knotted straps are clearly visible, takes two relaxed steps, makes a soft side turn and returns to camera in an easy hand-on-hip pose. Smiling directly into the lens, she says: “And the sandals make it feel so put together. Yeah, I’m wearing this everywhere.” End on a stable full-outfit pose long enough to read the bow graphics, sage straps, drawstring, pockets, layered ruffles and both floral sandals.
Use only the exact dialogue provided and synchronize mouth movement, gaze, expression and gestures to every line. Audio consists of clean creator speech, subtle room tone, light cotton rustle, drawstring movement and soft sandal footsteps; no music. Preserve the presenter’s identity and high ponytail, the ivory-and-sage palette, garment categories, bow and star graphics, camisole straps, drawstring, pocket placement, layered ruffled hems and floral sandal knots across all cuts. Prevent identity drift, hairstyle changes, wardrobe morphing, disappearing pieces, unstable graphics, missing pockets, fused fingers, duplicated limbs, clothing-body clipping, shorts becoming a skirt, sandals changing into other footwear, malformed feet, mirror errors, extra people, captions, subtitles, overlay text, brand logos, interface elements, borders and watermarks.
