For the target video, at 0.00 seconds into the target video, <Picture 1> (from [Shot 1]) is fully referenced. integrated_multimodal_description: [Shot 1] Live-action, vertical 9:16, warm indoor light—a medium already locked to <Picture 1>: the nude woman (S1) already kneels on the patterned rug, long dark hair, looking up, tongue already out; the man’s (S2) erect penis already in the foreground, her tongue already on the glans, circling it slowly. Wooden door left, bright window wash, white pillow as in the still. Her face, breasts, and kneeling body stay readable. No spoken lines. The camera holds with tiny amplitude at slow speed. [Shot 2] At 00:03.200, the camera cuts to a medium close-up of her mouth, tongue, and the shaft: she keeps circling the glans, then slowly licks down the underside toward the base, saliva shining. Her eyes and breasts stay in the top of frame, never a crop of only genitals. Not a room-wide. The handheld frame stays tight with tiny amplitude at slow speed. [Shot 3] At 00:07.000, the camera returns to a medium of her full kneeling figure and his hips and holds to 00:08.000: tongue at the base, still looking up, still on the rug—never a hall-wide. No on-screen captions, titles, subtitles, or extra lettering. overall_soundscape: Wet tongue on skin, her breath, his quiet exhale, room tone. No burned-in text. non_diegetic_music: none.
Parameters used to generate this content
AI models used to generate this content