Back 中文 CC Prompt Galaxy

Bystander handheld video: phone call misunderstanding AI Image Prompt

Prompt

Unprocessed iPhone handheld real video footage style, all camera settings auto, no post-production color grading or effects. Footage has realistic handheld micro-jitters and the operator's breathing sense, with natural lens dip or tilt from walking then quick recovery. Autofocus occasionally hunts, lags, and briefly loses focus, especially when following motion or hand movement. Auto white balance naturally shifts between warm and cool tones with sunlight and tree shadows. Image overall flat, preserving real lens flare, slight chromatic aberration at edges, motion blur and other optical imperfections. Mimics CCD sensor look with slight vertical smear, color bleed, grain noise, and purple/green fringing on high-contrast edges. Only in-frame natural ambient sound including street traffic, footsteps, wind in leaves; microphone may distort slightly during speech. Shot from street third-person bystander perspective (nearby passerby handheld), camera movement entirely follows bystander instinct (slight tracking, head turns). Characters 100% retain reference image model age, facial features, bone structure, eye spacing, jaw geometry, temperament, skin tone, hairstyle. No skin smoothing, no excessive softening, no averaging of facial features. 15-second timeline (complete scenario: phone call misunderstanding → woman mistakenly thinks she's being talked about → awkward turn → another misunderstanding reversal, third-person bystander perspective): 0s: Camera shakes slightly from third-person bystander perspective, framing a golden sycamore tree-lined street: a slender long-haired woman (wearing a mask, dressed as in image_20260608230651_1029_183(1)) walks toward camera with model-like gait, to the right stands a man in a black jacket (hf_20260612_104941_0727cb10-f615-43bf-8c49-2202835900bc) on the phone. Natural traffic and pedestrian ambience. Autofocus locks on their interaction, slight overexposure, CCD grain noise visible. 1-3s: Woman continues approaching; man speaks loudly into phone 'You dare to go out dressed like that?' Woman abruptly stops, turns sideways to glare angrily at man. Man looks innocent, shows phone call screen. Bystander camera slightly follows; autofocus briefly hunts then pulls back; CCD vertical smear noticeable. Mic captures fabric rustle and man's speech with slight distortion. 4-7s: Woman appears embarrassed, waves hand apologetically, responds 'Oh, sorry' and continues walking with head down. Man turns back to phone, continues talking loudly. Bystander camera naturally follows woman; slight shake and imperfect framing from walking; auto white balance drifts with tree shadows. Mic captures high-heel clicks and man's continued words 'Going out so scantily clad all day, what's the world coming to!' 8-11s: Woman walks a few steps then suddenly turns, strides back to man, raises hand as if to hit him. Man again bewildered, raises phone to show; both freeze in frame center. Bystander camera quickly pans; autofocus noticeably lags; background pedestrians briefly block view; handheld shake intensifies; CCD smear and grain persist. Mic captures woman's angry 'You're still saying you didn't talk about me!' and man's helpless explanation 'Miss, I'm really on the phone', with slight comedic ambiance. 12-15s: Woman finally awkwardly waves and leaves; man continues phone call. Bystander camera dips slightly then rises (weight shift), imperfect framing; woman's lower body briefly cropped at edge; faint fingerprint smudge marks appear at frame edges. Auto white balance shifts again; street ambience continues; overall real unprocessed handheld video texture.

AI Image Content Analysis

Content: Structure: Subjects (masked woman, black-jacket man), scene (sycamore street), style (iPhone handheld video, CCD effect), composition (third-person bystander perspective), lighting (auto white balance shifts), color (warm-cool transitions), texture (CCD grain, smear, chromatic aberration), lens (handheld shake, autofocus hunting), ratio (15-second timeline with staged description).

Pros: Rich details, clear timeline, high controllability; strong realism covering optical imperfections and handheld motion.

Cons: Prompt may be too long for model comprehension; some descriptions (e.g., 'faint fingerprint smudge') overly specific, increasing difficulty.

Reference image: Clearly depends on reference images (model, man), multiple reference image IDs.

Author
Unknown author
Type
Video
Aspect ratio
9:16
Size
720 × 1280
Added
2026-06-19

Source materials are collected from publicly accessible web pages. Contact us if a rights issue needs review. Copyright & Privacy