X精选 W16 · 城市人潮长曝光肖像 · 视频

Novel scene prompt:
{{selected_text}}

Create a short video generation request by adapting the cleaned X/Twitter visual style below to the novel scene.
Use one clear camera movement and one clear subject action. Preserve the novel characters and scene intent, but keep all people fictional adults unless the user explicitly provides authorized adult references.
Avoid public-figure likeness, real brands, sports teams, franchise IP, logos, living-artist names, and strong reference-preservation instructions.

Clean reusable visual style:
Professional high-fidelity transfer of the source X/Twitter prompt. Preserve its detailed design logic, camera language, composition hierarchy, lighting plan, material treatment, rendering quality notes, and negative constraints while adapting the subject, action, location, and story beat to the novel scene. Primary subject: fictional adult woman, 25+, original invented character, no public-person likeness.

Sanitized source specification:
Ultra-realistic cinematic street portrait of a fictional adult woman, 25+ standing still in a crowded city street, sharp focus on her face with calm, intense expression. Long-exposure effect with people moving around her creating strong directional motion blur, streaked crowd movement while the subject remains perfectly still. Natural soft daylight, shallow depth of field, creamy bokeh, realistic skin texture, subtle freckles, neutral makeup, dark winter coat and scarf. Emotional, introspective mood, urban storytelling photography, DSLR quality, 85mm lens look, f/1.8, high dynamic range, 8K, professional color grading.

Transfer requirements:
- Keep the original prompt's professional structure and specificity; do not collapse it into a short style summary.
- Replace any source-specific product, team, platform, franchise, or celebrity cue with a generic fictional equivalent.
- If reference imagery is used, it must be an authorized adult reference; otherwise generate an original fictional character.
- Preserve readable design intent such as lens choice, aspect ratio, depth layering, pose logic, lighting, texture, mood, and output goal when present.
- No real logos, no real brand marks, no public-person likeness, no childlike features, no explicit nudity, no watermark.

Video instructions:
Duration: {{duration_seconds}} seconds.
Aspect ratio: {{aspect_ratio}}.
Camera: choose a single cinematic movement that fits the scene, such as slow dolly-in, handheld street drift, gentle orbit, or locked-off portrait hold.
Motion: keep the main face and body stable; animate hair, fabric, light, crowd, smoke, powder, reflections, or background motion according to the style.
Reference notes: {{reference_notes}}.
Negative: flicker, morphing face, public figure likeness, childlike features, explicit nudity, extra limbs, warped hands, unreadable text, brand logo, franchise character, character drift across frames.