I2VA: For the target video, at 0.00 seconds into the target video, <Picture 1> is fully referenced as the exact starting frame and character identity. SUBJECT: Exact same freckled elf woman from Picture 1 — long white-blonde hair, pointed ears, green eyes, sheer white blouse. Maintain perfect facial and body likeness. ACTION / TIMING (9–10 seconds total): [0.0–3.5s] She stands looking slightly past the camera, then gets down on her hands and knees, searching the ground for her lost diamond. She completely ignores the viewer. [3.5–7.0s] As she leans forward on all fours the camera angle clearly shows down the front of her sheer blouse at her breasts while she continues looking at the ground. [7.0–10.0s] She slowly crawls a little closer toward the camera still searching, eyes focused downward, still ignoring the viewer. End on this. ENVIRONMENT / LOOK: Exact soft forest setting from Picture 1 — green foliage, dappled sunlight. CAMERA: Starts matching the medium reference framing. As she kneels, shifts to a higher angle looking down the front of her blouse to clearly show her breasts, then follows her as she moves closer. AUDIO (native stereo): - Soft movement and fabric sounds as she kneels and crawls - Quiet forest ambience (birds, light wind) - No music, no dialogue PRESERVE: Exact face, freckles, white hair, pointed ears, sheer blouse, and likeness from Picture 1. Keep the sequence: gets on hands and knees looking for the diamond → clear view of her breasts → crawls closer while still ignoring the viewer. AVOID: Having her look at or acknowledge the camera.
Parameters used to generate this content
The node graph used to generate this content — pan and zoom to explore, or download it to run in ComfyUI.
AI models used to generate this content