Between Us Local Short

subject_definitions: <Subject 1> is the adult woman in the first reference image, with a short black bob, a white sleeveless top, a gray skirt, and an ID lanyard. Preserve her facial features and overall appearance. <Subject 2> is the adult woman in the second reference image, with long pale-blonde hair, pointed elf-like ears, a pearl hair clip, a black sweater, a white collar, and a dark pleated skirt. Preserve her facial features and overall appearance. @Image1 is the third reference image, used as a shot-planning reference for the close interaction, lip-color touch-up, and intimate spacing between the two characters.

summary: [reference generation] Create a 15-second Ref2VA vlog-style video featuring <Subject 1> and <Subject 2>. Use @Image1 as an interaction and framing reference while adapting it to a single continuous handheld medium close-up with a Dutch angle. The characters remain absorbed in their private moment and never acknowledge the camera.

retention_analysis: <Subject 1> (appears throughout [Shot 1]): partially_preserved - retain her face, short black bob, white sleeveless top, gray skirt, and ID lanyard; adapt her pose and action to the new scene. <Subject 2> (appears throughout [Shot 1]): partially_preserved - retain her face, long pale-blonde hair, elf-like ears, pearl hair clip, and black-and-white outfit; adapt her pose and action to the new scene. @Image1 ([Shot 1] interaction and framing reference): partially_preserved - retain the intimate distance and lip-color touch-up action while using a medium close-up, a Dutch angle, and continuous handheld camera movement.

detailed_description: A naturalistic, intimate 15-second vlog-style moment, filmed as one uninterrupted take. [Shot 1] The shot opens in a softly lit room beside a bright window. In a medium close-up, <Subject 2> stands on the left and <Subject 1> leans in from the right. The camera is handheld at close conversational distance and held at a clear Dutch angle; gentle hand sway and small framing corrections keep the image feeling casually observed. Neither woman looks into the lens or reacts to it.

Following the interaction in @Image1, <Subject 1>, a quiet woman with a low, gentle voice (S1), brings a lipstick applicator toward <Subject 2>'s lower lip and says softly, <d>[Japanese] 動かないで。</d> <Subject 2>, with a light, breathy voice (S2), keeps her gaze lowered toward the applicator and replies, <d>[Japanese] 動いてないよ。</d> Their faces remain close. The camera drifts subtly closer, then steadies without cutting or changing shot size.

<Subject 1> applies the color with one small, deliberate stroke. <Subject 2>'s lips part slightly; she stays focused on <Subject 1>, not the camera. After the stroke, <Subject 1> lowers the applicator but does not immediately move away. A brief pause hangs between them. <Subject 2> looks up at <Subject 1> and murmurs, <d>[Japanese] ……近いね。</d> <Subject 1> holds her gaze for a beat, then quietly answers, <d>[Japanese] そう？</d> The handheld camera makes a small, imperfect lateral sway and settles on their faces, preserving the tilted composition through the end. Keep their faces, hairstyles, clothing, and spatial positions consistent throughout. No cuts, no camera-aware behavior, and no visible subtitles or added text.

overall_soundscape: Soft room tone and faint outdoor ambience filter through the nearby window. Quiet clothing movement and subtle handling of the lipstick applicator sit beneath the women's low, intimate voices.

non_diegetic_music: N/A