Process

Tools

Generation data

Prompt

External Generator
comfy
subject_definitions: <Subject 1> is the bear man in <Picture 1>, preserving his large brown bear build, thick dark fur, grey muzzle, black eyes, red pilot cap, and the beige short-sleeved utility shirt with rolled cuffs, buttoned placket, and chest pockets. In this scene his performance is easy and jovial: he is flying a plane, then then turns to the viewer to speak. <Picture 1> is the reference image defining <Subject 1>'s face, fur, and shirt. Use this for the background of the airplane cockpit. <Picture 2> is the reference image defining <Subject 1>'s face, fur, and shirt and hat. <Picture 3> is the reference image defining <Subject 1>'s relaxed posture with his hands behind his head. <Audio 1> is the voice of <Subject 1>, preserving his voice timbre, speaking rate, and the specific, suggestive content of his dialogue. summary: [keyframe completion + reference generation] Generate a twelve-second 1120x768 video that opens on <Subject 1> sitting in an airplane cockpit, flying the plane as clouds go by in the window from left to right. He slowly turns to the viewer and speaks. retention_analysis: <Subject 1> (appears in [Shot 1]): fully_preserved — hold the same face, fur, beige shirt, and confident expression. <Picture 1> ([Shot 1] reference): fully_preserved — the face, hat and shirt in the first frame. The background of the airplane cockpit. <Picture 3> ([Shot 1] reference): fully_preserved — posture and expression of <Subject 1>. detailed_description: A video in Disney animated style in 1120x768. [Shot 1] A medium shot frames <Subject 1> from the side as sits in the cockpit with his hands on the flight yoke. Clouds go by through the window behind him from left to right. The camera holds a static shot as he takes his hands off of the flight yoke and turns his head slightly to look over at the viewer. <Subject 1> (S1) says in a jovial southern accent: <d>[English] OK little britches. The Sea Duck's in auto pilot.</d> <Subject 1> swivels his chair to face the viewer. He slowly leans back and looks at the viewer. <Subject 1> (S1) says with an amused southern voice: <d>[English] We have some time to kill.</d> <Subject 1> places his hands behind his head. The camera pushes in with small amplitude at slow speed, focusing on <Subject 1>'s face. He is relaxed and has a slightly seductive expression on his face. <Subject 1> tilts his head slightly to the side. <Subject 1> (S1) says with a seductive voice: <d>[English] Maybe you can rustle up a snack for old Baloo, huh?</d>. The camera moves up to focus on <Subject 1>'s face as he looks seductively at the viewer. overall_soundscape: The low hum of an airplane engine and the sound of metal parts moving in [Shot 1]. non_diegetic_music: N/A

Other metadata

steps:12
sampler:Euler
engine:ComfyUI
denoise:1
scheduler:simple
comfy:

Discussion