Loop Forge

Selfie without stopping, shot 4

From MiniMax H3 long takes from chained renders

Frame from the Selfie without stopping, shot 4 clip
Watch it in the video at 2:05
Resolution
864×480
Length
192 frames, 8.00s
Handoff
Opens on the last 39 frames of the render before it
Steps
20
Seed
1078725334566040
Ran on
RTX 3060, 12GB (local)

What was wired in

  • Block 4
    <Picture 1>
    Block 4Aerial photograph of that block
  • Amber
    <Picture 2>
    AmberCharacter plate
  • Allie
    <Picture 3>
    AllieCharacter plate
  • No tagMiniMaxH3AddGuide: the last 39 frames of the previous render, plus its audio

The prompt

subject_definitions:
<Subject 1> is the young woman in <Picture 2>, in her early twenties, with long thick copper-orange curly hair falling well past her shoulders, pale freckled skin, blue-grey eyes and a slim build, wearing a plain cream short-sleeved crew-neck t-shirt, a long dark iron cross pendant on a thin dark chain, blue denim jeans, a wide tan leather belt with a brass buckle, a small tan leather belt pouch on her right hip, dark braided rope bracelets on both wrists and thin silver rings. Only her face, hair, body and clothing are taken from <Picture 2>; the snow, cliffs, green aurora and wooden sailing ship behind her in that image are not part of the target video.
<Subject 2> is the block of Market Street in San Francisco photographed from directly overhead in <Picture 1>: a dark red-brown triangular flatiron tower with a rounded white crown standing on the right side of the intersection, flanked by a pale grey stone tower and sheer dark glass high-rises, leafy street trees filling a landscaped median, brick-red paved sidewalks and several wide white-striped crosswalks meeting at the junction. Only the street layout, building shapes, materials and colours are used; the overhead aerial viewpoint and map-like flattening of <Picture 1> are not retained, and the target video is filmed at street level.
<Subject 3> is the young woman in <Picture 3>, in her mid-twenties, with long wavy ash-blonde hair, fair freckled skin, blue eyes and round thin gold-rimmed glasses, wearing a plain white spaghetti-strap slip dress and small stud earrings.

summary:
[reference generation + keyframe completion] The target video opens on an exact replay of the closing 1.6 seconds of the preceding segment of the same unbroken take and carries straight on from it as <Subject 1> passes the triangular flatiron tower of <Subject 2>, where <Subject 3> steps in beside her, takes a phone selfie of the two of them without either of them stopping, and then drops away to the left.

retention_analysis:
<Subject 1> (appears in [Shot 1]): fully_preserved - her face, freckles, blue-grey eyes, long copper-orange curls, cream t-shirt, iron cross pendant, tan leather belt and belt pouch, denim jeans, braided rope bracelets and silver rings are retained exactly.
<Subject 2> (appears in [Shot 1]): partially_preserved - the dark red-brown triangular flatiron tower with its rounded white crown, the pale grey stone tower, the dark glass high-rises, the leafy landscaped median, the brick-red paving and the wide striped crosswalks are retained; the overhead aerial viewpoint is replaced by a street-level camera at chest height.
<Subject 3> (appears in [Shot 1]): fully_preserved - her long wavy ash-blonde hair, fair freckled skin, blue eyes, round thin gold-rimmed glasses and plain white spaghetti-strap slip dress are retained.

detailed_description:
The target video is photorealistic live-action in a naturalistic documentary-cinematic style, filmed on a stabilised gimbal with a 35mm-equivalent lens held at chest height, in clear late-afternoon San Francisco light, with no on-screen text and no subtitles anywhere.
[Shot 1] The segment opens on an exact replay of the closing moments of the preceding take and carries that motion straight through with no join: the camera is already in front of <Subject 1>, facing back toward her and travelling backwards at exactly her walking speed, holding her centred and framed from mid-thigh up for the whole eight seconds with no cut. The camera holds its distance exactly for the entire segment and never pushes in, pulls out or zooms, so her height in the frame at the end is identical to her height in the frame at the start. She holds the exact body pose, head angle, gaze direction and facial expression carried in by the replayed opening frames, continuing them without any reset before moving into the actions below. <Subject 1>, the young woman with long copper-orange curls, the cream crew-neck t-shirt, the iron cross pendant, the tan leather belt with its small pouch and blue denim jeans, walks on toward the lens on the same foot and at the same unhurried cadence she carries out of the replayed opening. Behind her the junction of <Subject 2> opens up, the dark red-brown triangular flatiron tower with its rounded white crown standing tall over her right shoulder, the pale grey stone tower and dark glass high-rises stacked behind it, and the leafy landscaped median running away up the avenue. The low sun holds behind her, rimming her curls in warm orange and flaring gently into the lens. At about the two-and-a-half-second mark <Subject 3>, the young woman with long wavy ash-blonde hair, round thin gold-rimmed glasses and a plain white slip dress, pushes off from a tree in the median on the left and steps into frame beside her, matching her pace shoulder to shoulder. <Subject 1> (S1), her face turned square to the lens in clear three-quarter view with her mouth fully visible, her delight showing plainly in her eyes, raised brows and lifted cheeks rather than in her mouth, which stays free to work so her jaw and lips open and close distinctly on each syllable instead of settling into a fixed open smile or a laugh, and her hands staying down at her sides while she speaks, says in a bright, clear young female voice, <d>[English] Allie!</d> Only as the line ends does her face break into a delighted grin. <Subject 3> (S2), her face turned square to the lens in clear three-quarter view with her mouth fully visible, her delight showing plainly in her eyes, raised brows and lifted cheeks rather than in her mouth, which stays free to work so her jaw and lips open and close distinctly on each syllable instead of settling into a fixed open smile or a laugh, and her hands staying down at her sides while she speaks, replies in a soft, slightly breathy young female voice with a quick warm cadence, <d>[English] One quick one, keep walking!</d> Both women keep striding forward without slowing. Around the six-second mark, with both lines finished, they tilt their heads together, <Subject 1>'s copper curls falling against <Subject 3>'s ash-blonde hair, both smiling up into the raised phone; the phone screen brightens for an instant as the picture is taken. Only once she has finished speaking does <Subject 3> lift a dark phone up and forward at arm's length toward the camera side. <Subject 3> pulls the phone back down, glances at the screen, laughs and gives a small wave without looking up, then peels off toward the left and out of frame. <Subject 1> is left centred and alone again, still walking, tucking a strand of hair back behind her ear. In the final second the framing is unchanged and she walks straight into the lens, mid-stride, at the same steady pace.

overall_soundscape:
The open-air street ambience carries straight on from the preceding segment without interruption: traffic crossing the junction, wind in the median trees and the murmur of pedestrians. A second set of lighter footsteps joins hers for several seconds, a short synthetic camera-shutter click sounds from the phone near the middle of the segment, followed by a brief laugh that fades off to the left, leaving her even footsteps on the brick paving continuing alone.

non_diegetic_music:
N/A

Notes

A selfie taken without either of them stopping