
Seedance 2.0
PortraitSeedance 2.0
Tamsui sunset sweet girl moment
Creator @msjiaozhu
Prompt
15 seconds vertical screen 9:16, film feel filter + natural grain, Taipei Tamsui River bank walking path, background golden sunset + sparkling river surface + silhouette of Lover's Bridge in distance, main character Taiwan girl [Image 1], long hair slightly curly, beige knitted cardigan, clear makeup, throughout boyfriend Handheld POV follow shot, lens slight shake, no switching perspective, 0-4s: Handheld follow shot, girl walks two steps in front of lens, turns head to lens, brows lightly furrowed pretending to be angry, one hand pressing down hair messy from wind, other hand making a "keep up" gesture behind body, sunset side backlight outlines cheek contour and floating hair tips; Dialogue (soft coquettish): "Hey walk faster! You are dragging your feet!" 4-8s: Girl suddenly stops, looks at river surface, lens follows her gaze Smooth Pan panning to opposite bank -- golden sunset panorama + sparkling waves + Lover's Bridge silhouette, stays for about 2 seconds; Pan back to girl front, Golden Hour warm light dyes her face golden, she smiles eyes squinting into slits; Dialogue (short exclamation): "Wow—look at this light!" 8-10s: Girl's eyes light up looking at roadside stall, pulls lens direction walking fast, lens Handheld follows accelerating shake; [Natural street ambient sound, no dialogue] 10-13s: Girl already holding a tall Tamsui soft serve ice cream, holds it in front of lens pretending to feed; Dialogue (playful): "First bite for you—" pause one beat, she turns ice cream to herself opens mouth bites a big mouthful, because ice cream is too tall, nose tip rubs against cream top getting a bit of milky white when biting; Dialogue (smug): "--just kidding!" 13-15s: She chews ice cream, lowers head sees nose tip cream, cross-eyed glances down, freezes for half a second then laughs at lens, Handheld slight shake; Dialogue (laughing incomplete): "Hahaha... today you pass!" No text, subtitles, LOGO or watermark allowed.
Copy the text, add your own details, and try it in your image model.