Process

Tools

Techniques

  • img2vid

Generation data

Prompt

External Generator
comfy
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] Anime illustration, one continuous shot, no cuts. Hold the snow-leopard mage (S1) exactly as framed in <Picture 1>: white hair, cat ears, short pale muzzle, green half-lidded eyes, flat closed mouth, red earrings, striped shirt, white jacket and capelet with gold trim, white skirt, black tights, brown boots, spotted tail, and the crescent staff with the red gem in her right hand. The snowy pine forest and falling snow stay behind her. For the first half-second she is still, mouth closed, looking at the viewer. She plants the staff a little firmer. The red gem flares, and a thin ring of pale gold-red light leaves the crescent and travels straight toward the camera, growing as it approaches, lighting her face, jacket, and the snow at her boots. Her hair, capelet, and tail lift once in the spell wind, then settle. Her mouth stays sealed the whole time. She does not speak, does not smile, and does not blink into a new expression. The camera pushes in slowly with small amplitude and holds on the last beat as the light ring passes the lens. No second character. No elf ears. overall_soundscape: Cold wind in the pines, a soft hiss of falling snow, then a low rising magical hum as the gem flares and the light ring moves toward the camera. No voice, no breath-as-speech, no mumbling. non_diegetic_music: N/A

Other metadata

steps:4
sampler:res_multistep
seed:76651609
engine:ComfyUI
workflow:img2vid
denoise:1
duration:6
ecosystem:MiniMaxH3
scheduler:simple
comfy:

Discussion